by Jelle Christiaans
Humans not only experience the world through their senses but also have a strong propensity to communicate their inner world to other humans (“Mitteilungsbedürfnis”). Stories are one of the main ways we do this: we tell our children bedtime stories, pass down traditional stories about the origins of our families, and never stop telling our spouses how we ended up falling for them. As such, stories are a means to transfer a sensory experience to an audience and affect their senses. The narrator has different tools available to do this. One of these is words: when we hear that a ghost approaches the protagonist, a shiver runs down our back. Other tools include intonation, gestures, and facial expressions. Yet another is the focus of this article: grammar.
This article examines how speakers of Ecuadorian Siona (a Western Tukanoan language) use verbs to create a narratological experience for the audience—in particular, the grammatical category of verbal aspect. First, verbal aspect in general gives the audience a sense of time: it allows them to understand when and how events occurred in relation to each other. Second, verbal aspect in Ecuadorian Siona specifically can give the audience a sense of the structure of the story, specifically in the verb kaje “to say.” In particular, verbal aspect can inform the audience whether or not a particular discourse segment (usually a conversation) is already over. Third, verbal aspect in Ecuadorian Siona can be a device to instill a sense of suspense in the audience. These three functions will all be treated in succession in this article, but first I will elaborate on my corpus, methodology, and guiding theoretical assumptions.
Corpus and methodology
Ecuadorian Siona is spoken by roughly 250 people in Ecuador’s northeastern Sucumbíos province, where it is the predominant language in the towns of Soto Tsiaya and Puerto Bolívar. It is an endangered language because many children now grow up speaking Spanish instead. The language is related to Colombian Siona and varieties of Secoya, which are spoken in and around the same region.
The corpus of this study consists of two stories told by a native speaker of Ecuadorian Siona (henceforth: “Siona”). These were recorded by Martine Bruil, who also transcribed and translated them into Spanish and English in collaboration with native speakers. The first story is the hammock story (hãɨ̃dɨ in Siona) which is about a young man who, against the advice of the elders, goes to lie down in a hammock that turns out to be bewitched, and becomes stuck to the man’s back. The second story is the bat man story (ojobãĩ in Siona), which is about a cannibalistic father with some very strange habits who ends up eating his children and himself. Although ojobãĩ literally means “bat man,” it is used to mean “cannibal” in Siona.
The methodology consists of a careful close reading of these two stories, giving special attention to verb forms. This involves both a morphological and a semantic question. Not only is it important to establish which form a particular verb has (morphology) but also the discourse context that it is used in, from which it is possible to glean the meaning of the verb form (semantics). This close reading is coupled with two theoretical considerations outlined in the section below.
As Michael Halliday has recognized, the meanings of linguistic signs—which include both lexical categories (i.e., “words”) and grammatical categories—can lie on different communicative levels. I will here discuss two: the ideational level and the textual level. On the ideational level, linguistic signs have a meaning that describes a facet of the world of the language user or the world of a certain story (both subsumed under “narrated world”). For example, saying the word chicken refers to a chicken in a narrated world, and using the plural suffix -s signals that more than one chicken exists in the narrated world. But linguistic signs may also have a textual function, referring to facets of discourse itself. This becomes clear in example 1 below.
- Debby has many animals on her farm. First, she has a dozen chickens in her yard. Next, she keeps a couple pigs in her shed. Finally, two cats roam her property.
In this text, the words first, next, and finally do not (directly) say anything about the narrated world (i.e., about Debby’s farm or the animals on it); instead, they give the text structure. They help the audience understand not the content of what is being said but the way it is conveyed. As such, these linguistic signs refer to facets of discourse; they are concerned with the fabric of the text itself.
The German linguist Karl Bühler theorized the function of language and linguistic signs. He distinguished between sprachlicher Ausdruck (“linguistic expression”) and sprachlicher Appell (“linguistic appeal”). His point is that referentiality (sprachlicher Ausdruck)is only one function of the linguistic sign. That is, linguistic signs do more than refer to entities or events in a narrated world or facets of discourse. They may also be oriented toward influencing, affecting, and moving the audience. The clearest examples of this are imperatives such as Come!. These do not assert any event but impel the addressee(s) to carry out a particular action (in this case, to move toward the speaker).
Modern linguists now often refer to work by Roman Jakobson. He introduced English-speaking audiences to Bühler’s concept of “linguistic appeal” three decades after Bühler’s publication, also translating some of the German terminology into English. Since Jakobson, Bühler’s “linguistic appeal” has been known in English as the “conative function” of language, a term I will also use here.
The remainder of this article will use both the idea of the textual discourse function of linguistic signs and the conative function of languageto describe the use of verbal aspect in Ecuadorian Siona. Although much work on verbal aspect focuses on the ideational function, there has also been research on the interaction between discourse and aspect. It seems that for many languages, it is the type of discourse that affects the use of verbal aspect. This is the case for French, Dutch and English, as well as many Bantu languages. In all of these languages, the use of aspect depends on whether the language user is producing narrative, or if they are engaged in a non-narrative genre (often dialogues).
Paul Hopper showed that aspect can also make reference to discourse structure. In Russian, for instance, the imperfective can be used to signal digressions in discourse, or “non-sequencing” of events. In Classical Greek, the imperfective can be used to carry a narrative forward and describe its main storyline, whereas the perfective can introduce narrative intrusions. This article adds to the discussion of how aspect interacts with discourse structure by describing the particular way in which Siona verbal aspect gives narratives structure. I will also show how verbal aspect in Siona can have the conative function of language, which is something that has not been described before for verbal aspect in other languages.
Perfective and imperfective aspect
Although one can identify a number of different types of aspect, I will here only consider perfective and imperfective aspect. On the ideational level, the difference between these two is best understood as a distinction of completeness in temporal terms. For example, the English past simple generally expresses perfective aspect: the language user asserts that at a given moment in time, an event took place that also reached its end-point (example 2a). By contrast, the English past continuous expresses imperfective aspect: the language user asserts that at a given moment in time, an event took place that did not (yet) reach its end-point (example 2b).
(2) a. This morning, I biked to campus . . . and then
I rushed into my lab.
b. This morning, I was biking to campus . . . and
then got hit by a car.
In example 2a, the Past Simple biked informs the audience that the language user not only got on their bike and started making their way to campus but also that they reached that destination. As a result, the speaker can now start recounting events that happened after their arrival on campus (rushing into the lab). In 2b, however, the Past Continuous was biking informs the listener that the speaker got on their bike and started biking but had not yet reached campus. This gives the speaker room to relate events that happened during their journey to campus (e.g., getting hit by a car). It is important to note that imperfective aspect does not inform the audience that the end-point was never reached at all. The phrase I was biking only tells the audience that at this point in the story, the speaker is still on their way. One must infer from the rest of the story whether they made it at all.
Aspect forms in Siona
Siona encodes only perfective and imperfective aspect directly onto its verbs. Neither aspect is marked with a single form, and both use different endings depending on gender, number, and “type” of subject (see below). In tables 1a and 1b on the next pages, I have put the conjugations of verbal forms that are marked for aspect. It is important to note that Siona expresses verbal aspect only on non-final verb forms; verb forms at the end of a sentence are not marked for aspect.
As for number, aspect forms distinguish between singular and plural forms. As for gender, only the singular marks a distinction between masculine and feminine, whereas the plural is neutral with respect to gender. There is also a distinction between “same subject” and “different subject” forms: whereas saio indicates that the person who is on their way is also the person who is doing the next event (as in “when she was on her way, she passed by the river”), this is not the case for saiona, which indicates that the person who is on their way is not the same as the subject of the next event (as in “when she was on her way, I stayed home to cook”).
|Regular verb kaje “to say”|
|Aspect||Number/gender||Same subject||Different subject|
|Table 1a: Paradigms of aspectual verb forms in Siona.|
|Irregular verb saije “to go”|
|Aspect||Number/gender||Same subject||Different subject|
|Plural||sai-bɨ / sai-hɨ||sai-bɨ-na / sai-hɨ-na|
|Table 1b: Paradigms of aspectual verb forms in Siona.|
In these tables, I have not translated each form into English. The reason for this is that these forms do not correspond to a single sentence or phrase in English. For example, the verb saio can mean “when she was on her way, she . . .” but also “when I was on my way, I . . .” or “when you were on your way, you . . .” but only if the I or you is one woman (“feminine singular”). Moreover, the difference between “same subject” and “different subject” forms cannot be captured in a single translation, only transpiring in the context of language use.
A sense of time
Let us first consider the use of aspect on the ideational level in Siona. This is very similar to the situation in English that was described above. I here reiterate that on this ideational level, the difference between perfective and imperfective aspect is best understood as a distinction of completenessin temporal terms. With perfective aspect (perf), the language user asserts that at a given reference time (span), an event took place that also reached its end-point (example 3a). By contrast, imperfective aspect (impf) expresses that at a given moment in time, an event was taking place that did not (yet) reach its endpoint (example 3b).
(3) a. go’i-ni ĩ kɨa-ɨjã jehk-wa’i-de.
return-perf he told-m other-pl-acc
“After he returned he told the others.”
(3) b. ãi-hɨ tuhtu-hɨ-na jeo’ka-bi
eat-impf.pl sit.high-impf.pl-ds below-from
“While they were eating and sitting up top, something from below from the river moved and made these girls fall.”
In 3a, the perfective go’ini tells the listener that the male protagonist in the story not only started to make their way back to the village but that they also reached that destination at this point in the story (the reference point). As a result, the narrator can now start recounting events that happened after their return to the village (telling the other people about what had happened). In 3b, however, the imperfectives ãihɨ and tuhtuhɨna tell the listener that the girls were still eating and sitting at the reference time. This gives the narrator the room to relate events that happened during this eating and sitting (in this case, a mysterious object made them fall, putting an end to the eating and sitting). In these functions, verbal aspect in Siona gives the audience a sense of the temporal relations between events, similar to English.
A sense of structure
So far, I have shown how verbal aspect in Siona tells the audience something about the content of a story: it gives the audience a sense of how and when events happened in relation to each other. When we look at the verb kaje “to say,” however, we find a different situation. Instead of making reference to the way events unfold in the narrated world, verbal aspect in forms of kaje “to say” gives the audience a sense of the structure of the story they are listening to. More specifically, it does not indicate the completeness of a particular event but the completeness of a discourse segment as a whole (almost always a conversation). Imperfective aspect signals that the discourse segment has not yet come to an end, whereas perfective aspect signals that the discourse segment is over.
Let us look at an example of an imperfective form of kaje “to say” in example 4, below, which is from the bat man story. This sentence is from a passage where bat man exhibits very strange behavior, such as eating fish that is only half-cooked. When his wife comments on this peculiarity, bat man gives an unexpected reply.
(4) But when the fish was only half-cooked, he put it on a plate
and while he was eating it, his wife said “you are eating raw
ka-ko-na “zoe kwa’kusihko-a”
say-impf.sg.f-ds already cooked-copula
“When she had said (that), he said ‘it is already cooked’.”
Here, the imperfective form kakonarefers to a situation where bat man’s wife had finished speaking. As such, it expresses perfective aspect with respect to her “event” of speaking (her “speech act”)—that is, on Halliday’s ideational level. However, this form expresses imperfective aspect on the textual level of discourse: it signals that this particular conversation has not yet come to an end with the wife’s first remark in the previous line. It gives the audience a sense of anticipation: the wife’s remark stems from her confusion about bat man’s behavior, and she would like to have an explanation for it. Both the wife and the audience are waiting for this explanation, and thus, the imperfective form directs the attention of the audience to bat man’s reply that is to come. This interpretation is also supported by the suffix -na “different subject,” which indicates the subject of kakona (i.e., the wife) is different from the subject of kaɨjã (i.e., bat man).
There is a discrepancy between the aspect form in Siona and the English translation. In this case, the Siona imperfective form is not translated with an imperfective form in English but with a (past) perfective. This is because the Siona imperfective and the English imperfective have different ranges of functions. In English, the imperfective can only signal temporal overlap between events (ideational function), whereas the Siona imperfective can also reflect the structure of discourse, since it can anticipate a reply from the interlocutor in a conversation. In this case, it would be odd to translate kakona with an English imperfective, because (I argue) the Siona form kakona does not signal temporal overlap. A possible way to make the Siona meaning explicit in a paraphrase is “When she had said that, expecting her husband to reply, he said . . .”
If the Siona imperfective can signal the anticipation of a reply in a conversation, we find the opposite situation for perfective forms of kaje “to say.” An example of a perfective form is given in (5) below, which is from the hammock story. At this point in the story, the male protagonist has fled into the forest with the hammock stuck to his back. There, the hammock turns into a woman and orders the male protagonist to fulfill all of her wishes.
(5) “That other bunch, that big bunch of coconuts is ripe, take
that one down,” she said.
ka-o-na ĩ mɨ-ni kwã’sẽkɨwɨ say-perf.sg.f-ds he go.up-perf hook
nesihko-de mɨa-ni ĩ …
made-acc take-perf he …
“When she had said that, he went up and he took a hook
he had made and he . . .”
Here, we find the perfective form kaona,which refers to the speech act by the hammock in the previous line. Just like the imperfective kakonain (4), the perfective form refers to a situation where a speaker (in this case, the hammock) had started to speak and also finished what she had to say. In that respect, kaona and kakona are not different. However, the difference between them lies on the textual level of discourse: the perfective kaonaindicates that the conversation between the hammock and the male protagonist has come to an end. No reply from the male protagonist is anticipated, and the narrator proceeds to relate events that happen outside of the conversational context. In this case, the male protagonist uses one of his own hooks to grab coconuts from a nearby tree.
To summarize, verbal aspect in the verb kaje“to say” signals the completeness of a discourse segment (a conversation, by virtue of the meaning of kaje“to say”). Whereas imperfective aspect signals that the conversation has not yet come to an end and anticipates some kind of reply, perfective aspect signals that the conversation has reached its end and that something else, outside of the conversational context, is going to happen.
Sensing the suspense
Thus far, we have seen that imperfective forms of kaje“to say” are used to anticipate a reply from an interlocutor in a conversation. However, this is not the only thing that imperfective forms can anticipate in a story. They can also instill a sense of suspense in the audience by anticipating a particularly scary or gruesome moment. I will give two examples of this function in this section. The first is from the final passage of the hammock story, where the male protagonist has finally managed to shake off the bewitched hammock in the forest and repents his mistake in front of the elders back in the village.
(6) “ĩo jude mɨ’ hãɨ͂dɨ” ka-hɨ-na
she now your hammock say-impf.pl-ds
hẽõgõ-ni daɨ’ɨ” ka-kɨ ĩ
leave.behind-perf came say-impf.sg.m he
“meme-kɨna gwina-obi be.afraid-impf.sg.m do.again-nml.f
dah-si-o” ka-ni jowɨ
come-fut-f say-perf canoe
ti’wa-na ĩ hẽ-ni kãh-kɨjã.
side-on he cross-perf sleep-pst.m
“They said ‘and your hammock?’ and he said ‘she was entangled in a bunch of coconuts and I left her behind and came (here)’ and he said ‘I’m fearing she will come back and do it again’ and he went down into the canoe and he crossed it and fell asleep on the other side.”
The two boldfaced verb forms are of interest here, the imperfective kakɨ in particular. First, we find the imperfective form kahɨna,which indicates the first speech turn by the elders, who inquire about the absence of the hammock. They had seen the protagonist leave for the forest with the hammock and know that one cannot easily shake off a bewitched hammock; they would like to know how the male protagonist nonetheless managed to do this. The imperfective here functions as we saw in the previous section: it anticipates a reply by the male protagonist and indicates that the conversation is not yet over.
This reply then comes, and it is marked with the imperfective form kakɨ. However, no reply from the elders comes after this, and it turns out that kakɨ only interrupts the male protagonist’s final words in the story. The conversation is essentially over, and we might therefore have expected a perfective form to mark this end of the conversation instead. But we still find the imperfective form kakɨ in this case: why is this?
Here, the imperfective form kakɨ occurs at the end of the story and the final comment by the male protagonist is not without importance. The narrator has him end the story on an ominous note: even though the protagonist has shaken off the hammock for now, she is still roaming around, quite possibly waiting for revenge. Having the protagonist express his worry about this creates a big cliffhanger, and the imperfective form kakɨ creates the suspense that anticipates such a cliffhanger. As such, the imperfective fulfills the function of “linguistic appeal” (Karl Bühler) or the conative function of language (Roman Jakobson). It does not indicate how this particular discourse part is structured but instead aims at effectuating an emotion (i.e., fear and suspense) in the audience.
Thus far, we have only considered the verb kaje “to say” when looking at non-ideational functions of verbal aspect. Indeed, the division of labor seems clear: whereas verbal aspect in forms of kaje “to say” has the ability to structure discourse and create suspense, verbal aspect in other verbs functions on the ideational level, indicating if an event was completed or not yet at a particular point in the story. However, this division of labor is not a strict requirement of the grammar. Sometimes, other verbs may also be used to create suspense, as in example 7, which comes from the bat man story.
Previously in the story, bat man had—by way of ruse—invited his wife to look for their children near the river (recall that bat man had actually eaten their children but lied about it to his wife, and he intends to eat his wife too). When they go there, the wife goes to the river on her own for a moment, and she finds a round white object that turns out to be her son’s head. Horrified, she realizes that her husband is a cannibal (ojobãĩ in Siona, “bat man”) who intends to eat her as well, and she goes to hide up in a tree. She watches her husband from above as he calls her out to reveal herself and come to the campfire. At this point, the story reaches its narratological peak: the wife finds out about something even more gruesome than her husband’s cannibalism.
(7) She watched him, and he said “I am hungry” and took off
his tunic, the one he used to wear.
duhta-kɨ-na jã-ko-na hã-de
take.off-impf.sg.m-ds see-impf.sg.f-ds dem:dist-acc goa be’o-doro-wɨ-deba
ãi-sih-kɨ bah-kɨjã ĩ.
eat-compl-m.sg was-m he
“And when he had taken it off, she saw that he had cut off his private parts and he was all eaten.”
The most gruesome fact of this story has been revealed: bat man not only eats his own children, he eats at himself too. Compared to the revelation of the cannibalism, this auto-cannibalism instills an even greater sense of disgust in the audience, which is amplified by the suspense that the imperfect form duhtakɨna creates. In this case, it is difficult to determine if the imperfect has its ideational value: taking off a tunic is done within seconds, and it is only after one has taken off a piece of clothing that one can see what is underneath.
More pertinent here seems the conative function that the imperfect fulfills. It serves to create an anticipatory effect, and directs the attention of the audience to what happens next: upon hearing the imperfective, the audience knows that they are going to hear something horrendous, and the suspense has them sit at the edge of their seat. What will happen when bat man has taken off his tunic? What will he do, or what will be visible? In this function, the imperfective may be compared to the way ominous music is used in horror films to anticipate a jump scare. Without the music, you would still wonder what happens next (as one does throughout a film), but ominous music, like the imperfective in Siona, creates a peak of anticipation about the scary revelation to come.
This is different from the effect that a perfective form duhtaɨna would create. In that case, there would be no (special) anticipation about whatever it is that will happen after bat man’s has taken off its tunic. An audience may expect a more mundane and typical event happening after, such as bat man going to bathe in the river. This would be the equivalent of a lack of ominous music in a film.
In Ecuadorian Siona, the grammatical category of verbal aspect makes an important contribution to the sensory experience of a story by its audience. It not only gives the audience a sense of the temporal relation between events (the ideational function) but also gives the audience a sense of structure (the textual function): imperfective forms of kaje“to say” anticipate some type of reply in a conversation, whereas perfective forms signal that the conversation has reached its end. Narrators may also use the imperfective to instill a sense of suspense in the audience (sprachlicher Appell “linguistic appeal,” or conative function): in those cases, the imperfective does not signal that another conversational turn is expected but rather anticipates a particularly scary or gruesome event that is to come.
These different functions of verbal aspect in Ecuadorian Siona broaden our horizon of what grammar can do, particularly verbal aspect. Verbal aspect is not just a way to get the content of a message across intelligibly but also a narratological device that the narrator can use to interact with the audience and enhance the narratological experience for said audience. Ultimately, this study shows that grammatical categories must be taken seriously to fully understand how sensory experiences are constructed.
Allan, Rutger J. “The Imperfect Unbound. A Cognitive Linguistic Approach to Greek Aspect.” In Variation and Change in Ancient Greek Tense, Aspect and Modality,edited by Klaas Bentein, Mark Janse and Jorie Soltic, 100–30. Leiden: Brill, 2017.
Basset, Louis. “The use of the imperfect to express completed states of affairs. The imperfect as a marker of narrative cohesion.” In Discourse Cohesion in Ancient Greek, edited by Stéphanie J. Bakker and Gerry C. Wakker, 205–19. Leiden: Brill, 2010.
Bergqvist, Henrik and Karolina Grzech. “The Role of Pragmatics in the Definition of Evidentiality.” STUF Language Typology and Universals 76, no. 1 (2023): 1–30.
Bohnemeyer, Jürgen. “Aspect vs. Relative Tense: the Case Reopened.” Natural Language & Linguistic Theory 32, no. 3 (2014): 917–54.
Boogaart, Ronny. “Aspect and Temporal Ordering.” Ph.D. diss., Vrije Universiteit Amsterdam, 1999.
Bruil, Martine. Clause-typing and Evidentiality in Ecuadorian Siona. Utrecht: LOT Publications, 2014.
Bühler, Karl. “Die Axiomatik der Sprachwissenschaft.” Kant-Studien 38 (1933): 19–90.
Bybee, Joan L., Revere Perkins, & William Pagliuca. The Evolution of Grammar. Tense, Aspect an Modality in Languages of the world. Chicago/London: University of Chicago Press, 1994.
Carruthers, Janice. “Discourse and Text.” In The Oxford Handbook of Tense and Aspect, edited by Robert I. Binnick, 306–34. Oxford: Oxford University Press, 2012.
Comrie, Bernard. Aspect. Cambridge: Cambridge University Press, 1976.
ELAR. Accessed March 24, 2023. https://www.elararchive.org/.
Fleischman, Suzanne. “Discourse Functions of Tense-Aspect Oppositions in Narrative: Toward a Theory of Grounding.” Linguistics 23 (1985): 851–92.
—. Tense and Narrativity: from Medieval Performance to Modern Fiction. Austin: University of Texas Press, 1990.
Grønn, Atle. The Semantics and Pragmatics of the Russian Factual Imperfective. Ph.D. diss., Universitet i Oslo, 2003.
Gvozdanović, Jadranka. “Perfective and Imperfective Aspect.” In The Oxford Handbook of Tense and Aspect, edited by Robert I. Binnick, 781–802. Oxford: Oxford University Press, 2012.
Halliday, Michael A.K. An Introduction to Functional Grammar. London: Edward Arnold, 1985.
Harweg, Roland. “Aspekte als Zeitstufen und Zeitstufen als Aspekte.” Linguistics 181 (1976): 5–28.
Hoffmann, Thomas, and Graeme Trousdale. “Construction Grammar: Introduction.” In The Oxford Handbook of Construction Grammar, edited by Thomas Hoffmann and Graeme Trousdale, 1–12. Oxford: Oxford University Press, 2013.
Hopper, Paul J. “Aspect and Foregrounding in Discourse.” In Discourse and Syntax, edited by Talmy Givón, 213–41. Cambridge: Academic Press, 1979.
—. “Some observations on the typology of focus and aspect in narrative language.” Studies in Language 3 (1979): 37–64.
Jakobson, Roman. “Closing Statement: Linguistics and Poetics.” In Style in Language, edited by Thomas A. Sebeok, 350–449. Cambridge: M.I.T. Press, 1960.
Johnson, Marion R. “A Unified Temporal Theory of Tense and Aspect.” In Tense and Aspect (Syntax and Semantics 14), edited by Phillip Tedeschi & Annie Zaenen, 145–75. Paris: Brill, 1981.
Klein, Wolfgang. Time in Language. London/New York: Routledge, 1994.
Saussure, Ferdinand de. Cours de linguistique générale, edited by Charles Bally and Albert Sechehaye, assisted by Albert Riedlinger. Lausanne/Paris: Payot, 1916.
Swart, Henriëtte de. “Verbal Aspect.” In The Oxford Handbook of Tense and Aspect, edited by Robert I. Binnick, 752–80. Oxford: Oxford University Press, 2012.
Wal, Jenneke van der. Word Order and Information Structure in Makhuwa-Enahara. Utrecht: LOT Publications, 2009.
Zuberbühler, Klaus. “Syntax and Compositionality in Animal Communication.” Philosophical Transactions of the Royal Society B 375, no. 1789 (2020): 1–9.
 Zuberbühler, “Syntax and compositionality in animal communication,” 7.
 The root of this verb is ka- and -je is the ending for the infinitive, which is pronounced like the ge- in English, gem.
 Bruil, Ecuadorian Siona, 5.
 Here and throughout this article, I use the linguistic transcription of the language. Tildes on the vowel indicates a nasal vowel, and <j> sounds like English <j> as in jungle. The vowel [ɨ] is a close central unrounded vowel, which sounds similar to the pronunciation of e in glasses in most varieties of American English. The Siona also have their own orthography, where “hammock” is written ja̠ë̠rë and “bat man” is written oyoba̠i̠ for example. See Bruil, Ecuadorian Siona, 132, for a detailed comparison between these two orthographies.
 The hammock story is available in Bruil, Ecuadorian Siona, 343, and the bat man story is accessible through the online ELAR archive.
 The notion of the linguistic sign was famously first proposed by De Saussure as described by his students in Cours de linguistique Générale. It used to be a much more restricted notion, but nowadays, it has been extended in the way that I have presented it here. For the development of this notion, see Hoffmann and Trousdale, “Construction Grammar,” 1.
 Systemic Functional Grammar, xxvii.
 That is, a speaker, writer, or signer of a sign language.
 We might object that Halliday’s term textual function is biased towards the written word and is not cognizant of spoken discourse or sign languages, but I will leave this issue for now.
 Even though Bühler did not have the notion of a linguistic sign yet, we can well combine this theoretical notion with Bühler’s theory.
 Bühler, “Die Axiomatik der Sprachwissenschaft,” 81.
 “Closing Statement,” 355. Jakobson theorized even further than Bühler and described four additional functions of language. His six functions of language have become canonical within linguistics and are still widely used, e.g. Bergqvist and Grzech, “The Role of Pragmatics in the Definition of Evidentiality,” 4. For reasons of space, I will not discuss them further.
 E.g. the work that is still standard in linguistics, Comrie, Aspect. But see also De Swart, “Verbal aspect,” or Gvozdanović, “Perfective and Imperfective Aspect.”
 E.g. Fleischman, Tense and Narrativity.
 Boogaart, “Aspect and Temporal Ordering,” 224.
 E.g. Van der Wal, “Word Order and Information Structure in Makhuwa-Enahara,” 100.
 Hopper, “Aspect and Foreground in Discourse,” and Hopper “The Typology of Focus and Aspect in Narrative Language.” See also Fleischman, “Tense-Aspect Oppositions in Narrative.” There is also work on the use of the so-called “historical present” structuring discourse, but this concerns tense and not aspect, e.g. Carruthers, “Discourse and Text,” 307.
 Grønn, “The Russian Factual Imperfective,” 255–69.
 E.g. Basset, “The Use of the Imperfect,” or Allan, “The Imperfect Unbound.” A comparison between the findings for Classical Greek and my findings for Siona would be most worthwhile, but is much outside the scope of this paper.
 Bybee, Perkins and Pagliuca, The Evolution of Grammar; Bohnemeyer, “Aspect vs. Relative Tense,” 949.
 I consider aspect a temporal category, following Harweg, “Aspekt als Zeitstufe”; Johnson, “Tense and Aspect”; Klein, Time in Language.
 This end-point may be given by the verbal predicate itself in cases such as to walk to school (telic predicates) or by our knowledge of the world in cases such as to sleep (atelic predicates).
 Ecuadorian Siona has two verb classes, which Bruil calls non –i verbs and –i verbs. The former contains almost all verbs in the language and uses the same verb stem throughout the paradigm, but –i verbs are “irregular” in the sense that they have three verb stems that alternate in the paradigm, and for saije “to go” these are sai-, sah-, and sa-. The plural imperfective of the irregular verbs has the –bɨ suffix in the more conservative Soto Tsiaya variety, whereas the more innovative variety in Puerto Bolívar has taken the –hɨ suffix from the regular verbs. Bruil, Ecuadorian Siona, 209–10.
 Id., 175, 199. Bruil calls the former “dependent verbs” and the latter “main verbs.”
 Abbreviations used: 1 = first person, 2 = second person, 3 = third person, perf = perfective, impf = imperfective, sg = singular, pl = plural, m = masculine, f = feminine, nom = nominative, acc = accusative, neg = negation, pst = past tense, fut = future tense, dem = demonstrative, dist = distal, exist = existential, nml = nominaliser, cls = classifier, compl = completive, cont = container, intens = intensifier, ds = different subject (from the next verb). The vowel [ɨ] is a close central unrounded vowel, which sounds similar to the pronunciation of e in glasses in most varieties of American English. A tilde on a vowel indicates that the vowel is nasal. The apostrophe represents the glottal stop, which also occurs in the middle of English uh-oh. The <j> and <ch> sound like English <j> and <ch>as in jungle and chicken.
 Bruil, Ecuadorian Siona, 351, line 27. All translations of the Siona example sentences are mine.
 Id., 161. There is no overt subject expressed for the main verb, so that the subject is interpreted as an indefinite subject “something.”
 ELAR, Bat man story, line 18.
 Bruil, Ecuadorian Siona, 350, line 23.
 Id., 352, line 28.
 ELAR, Bat man story, line 49.