1 Introduction

This paper investigates the evidential markers dizque and como que in the American Spanish varieties spoken in Bogotá and Medellín, Colombia. Evidentiality is a grammatical category whose core meaning encodes the source of information an individual draws on regarding that piece of information (see Willett 1988; Izvorski 1997; Van der Auwera & Plungian 1998; Dendale & Tasmowski 2001; Plungian 2001; Faller 2002; 2012; 2019; Aikhenvald 2004; 2006; 2018; Matthewson et al. 2007; McCready 2015; Rooryck 2019). Evidentiality in Spanish specifically has been studied as well (see Kany 1994; Magaña 2005; Olbertz 2005; 2007; López Izquierdo 2006; Travis 2006; Cruschina & Remberger 2008; Babel 2009; Demonte & Fernández-Soriano 2013; 2020; 2022; Alcázar 2014; 2018; De la Mora & Maldonado 2015; Grajales 2017; Saito 2019; 2021; Martínez Vera 2020a; 2023; Sanromán Vilas 2020). Across varieties, the markers that have been studied more extensively are que, which appears to be very widespread, and dizque, which is widely attested in American Spanish.

Focusing on American Spanish, work over the past years has addressed evidentiality concentrating on different varieties. For instance, it has been reported that dizque is widely attested (see Sanromán Vilas 2020 for a recent corpus study). In some varieties, such as Mexican Spanish, dizque appears to covary with another element, i.e., quesque (De la Mora & Maldonado 2015). Recent work has further pointed out the extensive variation there is across varieties, including extensive speaker variation within a single variety (Martínez Vera 2023). To explore the extent of the variation further, this paper examines evidentiality in Colombian Spanish, in particular, it concentrates on two evidential markers that are attested in the varieties of Bogotá and Medellín, namely, dizque and como que, in matrix declarative clauses (see Grajales 2017; see also Camacho 2011). These are exemplified in (1).1 In the examples, dizque (1a) and como que (1b) appear at the beginning of the clause; the scope propositions is that it is going to rain tonight—the free translations attempt to capture the intuitive meaning of these sentences.2

    1. (1)
    1. a.
    1. Dizque
    2. dizque
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche.
    2. night
    1. ‘Allegedly, it is going to rain tonight.’
    1.  
    1. b.
    1. Como
    2. como
    1. que
    2. que
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche.
    2. night
    1. ‘It looks like it is going to rain tonight.’

The sentences in (1) differ with regard to the evidential meaning associated with them: (1a) is felicitously uttered in the presence of reportative evidence (e.g., the speaker was told about the rain), while (1b) requires inferential evidence (e.g., the speaker drew an inference about it raining based on the weather conditions). Furthermore, these sentences have an epistemic implication of lack of certainty (see De la Mora & Maldonado 2015). Thus, both sentences convey the meaning that the speaker expresses reservation (intuitively, the speaker expresses doubt) towards the proposition that is conveyed.

As indicated, the examples in (1) illustrate cases with dizque and como que in sentence-initial position. Although evidential marking occupies a fixed position in the clause in many languages, dizque and como que can appear in different positions in the sentence, as in the examples in (2). With regard to dizque, this has been referred to as the labeling function (Travis 2006), but it extends to como que. In addition to incorporating specific evidential requirements (see (1)), these sentences intuitively mean that the epistemic implication is closely tied to the constituent following the evidential. Thus, the speaker expresses lack of certainty in that she is unsure as to whether the relevant object is in fact a sea compass (this can be further seen by means of the proposed correction in parentheses).

    1. (2)
    1. a.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. dizque
    2. dizque
    1. un
    2. a
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought a kind of a sea compass (that ended up being a photo viewer).’
    1.  
    1. b.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. como
    2. como
    1. que
    2. que
    1. un
    2. a
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought a kind of a sea compass (that ended up being a photo viewer).’(see Grajales 2017)

This paper provides a unified account of cases where dizque and como que in Colombian Spanish take clausal scope and of cases involving the so-called labeling function (Travis 2006). We take into consideration (i) their evidential contribution and (ii) the epistemic implication of lack of certainty associated with these markers, in combination with (iii) their distribution as tied to the different positions they may occupy in the clause. We propose that dizque and como que are evidential markers that are also involved in focus marking (see Rooth 1985; 1992), i.e., dizque and como que are hybrid evidential and focus markers. To the best of our knowledge, the claim that (some) evidentials can simultaneously be focus markers has been made only to a limited extent. Specifically, such a claim has been made for the Andean languages Quechua and Aymara. Thus, Faller (2002) (following Muysken 1995), indicates that the Cuzco Quechua evidentials are also focus markers; only the evidential component is analyzed in this work. Sánchez (2010) proposes a syntactic analysis of the Cuzco Quechua evidentials that attempts to integrate the two functions. Similarly, Martínez Vera (2020a; 2024) proposes that the direct evidential =wa in Southern Aymara is also a focus marker (see also Proulx 1987) and proposes a semantic analysis.

This paper provides an analysis that concentrates on syntactic and semantic aspects of dizque and como que as hybrid evidential and focus markers. To this end, in addition to addressing the evidential contribution of cases with these markers, we liken them to focus-sensitive items (e.g., only; see Hirsch 2017; Rochemont 2018; Greenberg 2022 for recent work) and provide an initial comparison of the distribution of dizque and como que vs. solo ‘only’, a better-known focus-sensitive adverb in Spanish (Kovacci 1999; Sánchez López 1999). We also examine cases of question-answer congruence (Hamblin 1973; Kartunnen 1977; Groenendijk & Stokhof 1984; Rooth 1985; 1992; Krifka 1992; Ginzburg 1996; Roberts 1996), to which sentences with dizque and como que are sensitive. Specifically, these markers have to be adjacent to the focused constituent in question-answer pairs where the question manipulates the focus. We further make the novel claim that the epistemic implication of lack of certainty in the cases under consideration arises in connection to focus alternatives as determined by dizque and como que and the constituent in their scope.

This paper is organized as follows. Section 2 concentrates on the evidential contribution and the epistemic implication of lack of certainty of sentences with dizque and como que. Section 3 turns to the distribution of these markers with regard to their scope possibilities; an initial comparison with solo ‘only’ is established. Section 4 turns to question-answer congruence. Sections 3–4 suggest that dizque and como que are involved in focus marking. Section 5 provides a unified analysis of the cases under consideration; the proposal captures the distribution of dizque and como que, including the consideration of locality restrictions, as well as their evidential contribution and focus alternatives, which are connected to the epistemic implication of lack of certainty that arises.

2 Evidential contribution and epistemic implication

This section concentrates on the evidential contribution and the epistemic implication of lack of certainty of sentences with dizque and como que.3 Sentences with dizque are felicitously uttered when the speaker has reportative evidence for the relevant scope proposition (in the absence of such evidence, they are infelicitous). The reportative evidence can be varied in that it can be second- or third-hand (see Cruschina & Remberger 2008; see Martínez Vera 2023 for additional contexts). Thus, the sentences in (3) ((3b) repeats (1a)) are felicitously uttered when the speaker was told (by a known source) that it is raining now (3a) or that it would rain tonight (e.g., a work colleague of the speaker told her that it is going to rain tonight) (3b), but they are also felicitously uttered when the source is unknown (e.g., rumor has it that it is raining now or that it is going to rain tonight). Sentences with dizque are infelicitous in the presence of direct and inferential evidence. With regard to direct evidence, consider a scenario where the speaker sees that it is raining by looking outside; (3a) cannot be felicitously uttered in this scenario. With regard to inferential evidence, the sentences in (3) are infelicitous in scenarios such as the following: (3a) cannot be felicitously uttered if the speaker draws the inference that it is raining by seeing the wet coat of somebody that has just come in; (3b) cannot be felicitously uttered either if the speaker draws the inference that it would rain tonight because it is very cloudy, which normally means that it would rain in Medellín; (3b) is also infelicitous if uttered based on general knowledge, e.g., that it is the rainy season in Medellín.4,5

    1. (3)
    1. a.
    1. Dizque
    2. dizque
    1. llueve
    2. rains
    1. (ahora).
    2.   now
    1. ‘Allegedly, it is raining (now).’
    1.  
    1. b.
    1. Dizque
    2. dizque
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche.
    2. night
    1. ‘Allegedly, it is going to rain tonight.’

While similar to cases with dizque in that cases with como que are not compatible with direct evidence, dizque and como que behave differently in that only como que requires the presence of inferential evidence (in the absence of such evidence, they are infelicitous). Thus, the examples in (4) ((4b) repeats (1b)) can be felicitously uttered in a context in which the speaker draws the inference that it is raining now because someone came in with a wet coat (4a) or that it would rain tonight because it is very cloudy, which normally means that it would rain in Bogotá (4b). The sentences in (4) can also be uttered in a scenario in which, based on common knowledge about Bogotá’s weather (e.g., the speaker’s experience through the years, which is shared across the population), the speaker infers that it is raining now or that it would rain that night. In contrast, the sentences in (4) are infelicitous against the reportative scenarios indicated above for (3) in which dizque is felicitous; (4a) in particular is also infelicitous in the presence of direct evidence, as in the scenario indicated above where the speaker sees that it is raining by looking outside.

    1. (4)
    1. a.
    1. Como
    2. como
    1. que
    2. que
    1. llueve
    2. rains
    1. (ahora).
    2.   now
    1. ‘It looks like it is raining (now).’
    1.  
    1. b.
    1. Como
    2. como
    1. que
    2. que
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche.
    2. night
    1. ‘It looks like it rains/is going to rain tonight.’

The examples in (3)–(4) further involve an epistemic implication of lack of certainty (see Travis 2006; Alcázar 2018), i.e., the speaker expresses that she is unsure as to whether the scope proposition is true—intuitively, this means that the speaker distances herself with respect to what is uttered (see footnote 4). Thus, (3)–(4) are infelicitous if the speaker knows for sure that the scope proposition is true. In this regard, imagine a context where the speaker has no doubts whatsoever about the scope proposition. For dizque, suppose that the source is very reliable; the speaker has no reason to doubt the source’s report (see Martínez Vera 2023). For como que, suppose that the relevant state of affairs in Medellín with regard to the weather indicates that it is certain that it is already raining or that it will start to rain soon (e.g., with the clouds as they are, there is no other possible outcome). (3)–(4) are infelicitous in these scenarios. In contrast, if the speaker does not believe the source of the report (for dizque) or if the cloudy state of affairs in Medellín is not as conclusive as one would like (for como que), (3)–(4) can be felicitously uttered.

The epistemic implication of lack of certainty has been explicitly targeted for dizque in recent work (Martínez Vera 2023) by means of a follow-up of the form (yo) no tengo dudas al respecto ‘I don’t have any doubts about it,’ which targets whether the speaker is sure or not about the scope proposition that is uttered. The test shows that the follow-up is marginal, thus confirming the speaker’s lack of certainty with regard to the scope proposition when uttering it.6

    1. (5)
    1. a.
    1. Dizque
    2. dizque
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche;
    2. night
    1. ??(yo)
    2.      I
    1. no
    2. no
    1. tengo
    2. have
    1. dudas
    2. doubts
    1. al
    2. to.the
    1. respecto.
    2. respect
    1. ‘Allegedly, it is going to rain tonight; I don’t have any doubts about it.’
    1.  
    1. b.
    1. Como
    2. como
    1. que
    2. que
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche;
    2. night
    1. ??(yo)
    2.      I
    1. no
    2. no
    1. tengo
    2. have
    1. dudas
    2. doubts
    1. al
    2. to.the
    1. respecto.
    2. respect
    1. ‘It looks like it is going to rain tonight; I don’t have any doubts about it.’(see Grajales 2017)

While sentences with both dizque and como que involve lack of certainty, they differ with regard to whether they may further involve disbelief (see Stalnaker 1978; 2002; see also Faller 2002 for discussion about evidentials). Sentences with dizque are compatible with the speaker believing that the scope proposition is false. In contrast, in the case of como que, the speaker has to believe that the scope proposition is possible. To this end, we provide tests that involve continuations that indicate that the speaker does not believe the scope proposition or that the speaker is unsure about whether the state of affairs indicated by the scope proposition holds.7 In the presence of dizque (6a), the speaker may believe that the scope proposition is false; she may also remain undecided. In contrast, in the presence of como que (6b), the speaker may be undecided with respect to the scope proposition, but it is not possible for her to fully discard it—it is infelicitous for her to state that she does not believe the scope proposition. In section 5.2, we make a proposal that accounts for this distinction, which builds on AnderBois (2014) and Faller (2002; 2019) for dizque, and on Matthewson et al. (2007) and Faller (2002) for como que.

    1. (6)
    1. a.
    1. Dizque
    2. dizque
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche,
    2. night
    1. pero
    2. but
    1. yo
    2. I
    1. no
    2. no
    1. lo
    2. CL
    1. creo /
    2. believe
    1. no
    2. no
    1. estoy
    2. am
    1. seguro
    2. sure
    1. de
    2. of
    1. si
    2. if
    1. así
    2. like.that
    1. será.
    2. will.be
    1. ‘Allegedly, it is going to rain tonight, but I don’t believe it/I’m not sure whether it’ll actually rain.’
    1.  
    1. b.
    1. Como
    2. como
    1. que
    2. que
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche,
    2. night
    1. pero
    2. but
    1. #yo
    2.   I
    1. no
    2. no
    1. lo
    2. CL
    1. creo /
    2. believe
    1. no
    2. no
    1. estoy
    2. am
    1. seguro
    2. sure
    1. de
    2. of
    1. si
    2. if
    1. así
    2. like.that
    1. será.
    2. will.be
    1. ‘It looks like it is going to rain tonight, but I don’t believe it/I’m not sure whether it’ll actually rain.’

The examples in (3)–(6) have focused on cases where dizque and como que modify the whole clause. It is worth pointing out that the evidential contribution and epistemic implication of lack of certainty are also present when dizque and como que appear next to (and modify) a subclausal constituent. Here we illustrate this with an example concentrating on this aspect (see section 3 for more in-depth discussion). Thus, the examples in (7) (=(2)) are felicitously uttered if the speaker has reportative evidence for the proposition that Juan brought a sea compass (e.g., the speaker was told about it) (7a) and if the speaker infers that Juan brought a sea compass (e.g., the speaker infers that Juan brought a sea compass, because Juan is a sailing aficionado, and the speaker knows that he was missing a sea compass and was going to get things related to sailing soon) (7b).8

    1. (7)
    1. a.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. dizque
    2. dizque
    1. un
    2. a
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought a kind of a sea compass (that ended up being a photo viewer).’
    1.  
    1. b.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. como
    2. como
    1. que
    2. que
    1. un
    2. a
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought a kind of a sea compass (that ended up being a photo viewer).’(see Grajales 2017)

Furthermore, the epistemic implication of lack of certainty is also present in (7). It is tied to the constituent following dizque and como que, i.e., the speaker’s intended implication suggests that what is denoted by the expression adjacent to dizque and como que may come under scrutiny. Intuitively, this means in (7) that there is lack of certainty in the speaker as to whether the object that Juan brought is in fact a sea compass. That this is the case is targeted by means of the continuation that appears in parentheses in the examples: the speaker corrects herself by indicating the a posteriori finding or realization that the object was a different one. That this is the case can be further supported by an explicit continuation that involves lack of certainty concentrating on the element following dizque and como que. We target this below by means of a continuation that indicates that the speaker has no doubt that Juan brought something, but she is not sure if it really was a sea compass.

    1. (8)
    1. a.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. dizque
    2. dizque
    1. un
    2. a
    1. compás
    2. compass
    1. marino;
    2. sea
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. algo,
    2. something
    1. no
    2. no
    1. hay
    2. is
    1. duda,
    2. doubt
    1. pero
    2. but
    1. no
    2. no
    1. estoy
    2. am
    1. seguro
    2. sure
    1. de
    2. of
    1. si
    2. if
    1. fue
    2. was
    1. un
    2. a
    1. compás
    2. compass
    1. marino.
    2. sea
    1. ‘Juan brought a kind of a sea compass; Juan brought something, there’s no doubt about it, but I’m not sure whether it was a sea compass.’
    1.  
    1. b.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. como
    2. como
    1. que
    2. que
    1. un
    2. a
    1. compás
    2. compass
    1. marino;
    2. sea
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. algo,
    2. something
    1. no
    2. no
    1. hay
    2. is
    1. duda,
    2. doubt
    1. pero
    2. but
    1. no
    2. no
    1. estoy
    2. am
    1. seguro
    2. sure
    1. de
    2. of
    1. si
    2. if
    1. fue
    2. was
    1. un
    2. a
    1. compás
    2. compass
    1. marino.
    2. sea
    1. ‘Juan brought a kind of a sea compass; Juan brought something, there’s no doubt about it, but I’m not sure whether it was a sea compass.’

If lack of certainty arises in connection to a (subclausal) constituent as discussed, a natural question to entertain is whether the evidential contribution behaves similarly. In this regard, some evidentials have been argued to provide evidence for a constituent cross-linguistically, e.g., an evidential in a DP may provide the source of evidence the speaker has for the existence of the entity that such a DP denotes (Hanks 2009; Gutiérrez & Matthewson 2012; Reisinger et al. 2021). The cases under consideration, however, do not seem to behave this way with regard to the evidential contribution, which has to be of the whole proposition. Focusing on (7a), with dizque, suppose that the speaker has seen Juan bringing something, although she could not see what. Someone else then tells the speaker that they saw Juan carrying (what looked like) a sea compass. Thus, the speaker would have direct evidence that Juan is carrying something, but would only have reportative evidence for what that item might be. In such a case, (7a) is infelicitous. As for (7b), with como que, suppose again that the speaker has seen Juan bringing something, although she could not see what. Previously (e.g., a couple of days before), the speaker had heard that Juan was gathering things that are used to go sailing, and she further knows that sea compasses are things that Juan really likes, so she infers that Juan must have a sea compass with him. Thus, the speaker would have direct evidence that Juan is carrying something, but would only have inferential evidence for what that item might be. (7b), however, is infelicitous against this scenario. The discussion thus suggests that dizque and como que are felicitously uttered with regard to evidence when the speaker has reportative and inferential evidence respectively for the whole proposition.

This section has shown that sentences with dizque and como que are felicitously uttered in the presence of relevant evidence for the scope proposition, namely, reportative evidence in the case of dizque and inferential evidence in the case of como que. In addition, sentences with these evidentials further convey lack of certainty in the speaker towards the scope proposition—lack of certainty is tied to the constituent following the evidential. These two aspects, namely, the evidential contribution and the epistemic implication of lack of certainty must be present in cases in which both clauses and subclausal constituents appear next to the evidentials.

3 Positions in the clause and comparison with solo ‘only’

As indicated, dizque and como que may appear in different positions in the clause, specifically, they can combine with and thus modify different kinds of constituents (e.g., clauses, but also subclausal constituents), as illustrated in (1)–(2) respectively (see Assmann et al. 2023 for extensive discussion of this issue for several focus markers cross-linguistically; see section 4 for additional discussion). Furthermore, the evidential contribution and the epistemic implication of lack of certainty must also be present. This section provides additional examples of the different positions that these evidentials can occupy in the clause. While this section concentrates on the syntactic distribution of dizque and como que, it must be kept in mind that, in all the examples that follow, the relevant evidence must also be present (see section 2). In addition, there is an epistemic implication of lack of certainty, which is closely tied to the constituent that appears adjacent to the evidentials (see (7) above).

The evidentials dizque and como que frequently appear in sentence-initial position (1); in fact, they may precede the subject if it is overt, as shown in (9). In these cases, the evidentials combine with and modify the whole clause (including the subject) (see Martínez Vera 2023; see also López 2009). The markers may also appear between an overt subject and the verb (10), which are instances in which what is modified is a smaller constituent, e.g., the TP completó la primera vuelta in this case.9

    1. (9)
    1. a.
    1. Dizque
    2. dizque
    1. Milagros
    2. Milagros
    1. completó
    2. completed
    1. la
    2. the
    1. primera
    2. first
    1. vuelta.
    2. round
    1. ‘Allegedly, Milagros completed the first round.’
    1.  
    1. b.
    1. Como
    2. como
    1. que
    2. que
    1. Milagros
    2. Milagros
    1. completó
    2. completed
    1. la
    2. the
    1. primera
    2. first
    1. vuelta.
    2. round
    1. ‘It looks like Milagros completed the first round.’
    1. (10)
    1. a.
    1. Milagros
    2. Milagros
    1. dizque
    2. dizque
    1. completó
    2. completed
    1. la
    2. the
    1. primera
    2. first
    1. vuelta.
    2. round
    1. ‘Milagros allegedly completed the first round.’
    1.  
    1. b.
    1. Milagros
    2. Milagros
    1. como
    2. como
    1. que
    2. que
    1. completó
    2. completed
    1. la
    2. the
    1. primera
    2. first
    1. vuelta.
    2. round
    1. ‘Milagros appears to have completed the first round.’

Furthermore, dizque and como que frequently appear between an auxiliary and the verb (11) but can appear naturally preceding a variety of constituents such as DPs (12),10 PPs (13), and less frequently, adverbs (14).

    1. (11)
    1. a.
    1. La
    2. the
    1. lluvia
    2. rain
    1. sigue
    2. keeps
    1. dizque
    2. dizque
    1. cayendo.
    2. falling
    1. ‘The rain keeps allegedly falling.’
    1.  
    1. b.
    1. La
    2. the
    1. lluvia
    2. rain
    1. sigue
    2. keeps
    1. como
    2. como
    1. que
    2. que
    1. cayendo.
    2. falling
    1. ‘The rain keeps it looks like falling.’
    1. (12)
    1. a.
    1. Va a
    2. it.is.going.to
    1. llover
    2. rain
    1. dizque
    2. dizque
    1. esta
    2. this
    1. noche.
    2. night
    1. ‘It is going to rain allegedly tonight.’
    1.  
    1. b.
    1. Va a
    2. it.is.going.to
    1. llover
    2. rain
    1. como
    2. como
    1. que
    2. que
    1. esta
    2. this
    1. noche.
    2. night
    1. ‘It is going to rain it looks like tonight.’
    1. (13)
    1. a.
    1. Van a
    2. are.going.to
    1. llegar
    2. arrive
    1. dizque
    2. dizque
    1. a
    2. at
    1. las
    2. the
    1. nueve.
    2. nine
    1. ‘They are going to arrive allegedly at nine.’
    1.  
    1. b.
    1. Van a
    2. are.going.to
    1. llegar
    2. arrive
    1. como
    2. como
    1. que
    2. que
    1. a
    2. at
    1. las
    2. the
    1. nueve.
    2. nine
    1. ‘They are going to arrive it looks like at nine.’
    1. (14)
    1. a.
    1. El
    2. the
    1. joven
    2. youth
    1. escribió
    2. wrote
    1. dizque
    2. dizque
    1. siempre
    2. always
    1. Gucci,
    2. Gucci,
    1. nunca
    2. never
    1. “in-Gucci”.
    2. “in-Gucci”
    1. ‘The young man allegedly always wrote Gucci, never “in-Gucci”.’(Facebook 3/15/2020)
    1.  
    1. b.
    1. El
    2. the
    1. joven
    2. youth
    1. escribió
    2. wrote
    1. como
    2. como
    1. que
    2. que
    1. siempre
    2. always
    1. Gucci,
    2. Gucci,
    1. nunca
    2. never
    1. “in-Gucci”.
    2. “in-Gucci”
    1. ‘The young man it looks like always wrote Gucci, never “in-Gucci”.’

Constituent dizque and como que are also attested inside DPs and PPs. As shown in (15), dizque and como que modify the constituent to their right. Thus, these constituents are the PP (15a, b) and the NP (15c, d). Similarly, in the case of PPs, dizque and como que can modify the P’s complement, as exemplified in (16).11,12

    1. (15)
    1. a.
    1. Mira,
    2. look
    1. desde
    2. since
    1. secundaria
    2. highschool
    1. teníamos
    2. had
    1. un
    2. a
    1. grupo
    2. group
    1. dizque
    2. dizque
    1. de
    2. of
    1. beneficencia
    2. charity
    1. y
    2. and
    1. esas
    2. those
    1. cosas.
    2. things
    1. ‘Look, since highschool, we had an alleged charity group, and such.’(adapted from Olbertz 2007; see Grajales 2017)
    1.  
    1. b.
    1. Mira,
    2. look
    1. desde
    2. since
    1. secundaria
    2. highschool
    1. teníamos
    2. had
    1. un
    2. a
    1. grupo
    2. group
    1. como
    2. como
    1. que
    2. que
    1. de
    2. of
    1. beneficencia
    2. charity
    1. y
    2. and
    1. esas
    2. those
    1. cosas.
    2. things
    1. ‘Look, since highschool, we had it looks like a charity group, and such.’
    1.  
    1. c.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. un
    2. a
    1. dizque
    2. dizque
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought an alleged sea compass (that ended up being a photo viewer).’
    1.  
    1. d.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. un
    2. a
    1. como
    2. como
    1. que
    2. que
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought a kind of a sea compass (that ended up being a photo viewer).’
    1. (16)
    1. a.
    1. Van a
    2. are.going.to
    1. llegar
    2. arrive
    1. con
    2. with
    1. dizque
    2. dizque
    1. varios
    2. several
    1. amigos.
    2. friends
    1. ‘They are going to arrive allegedly with several friends.’
    1.  
    1. b.
    1. Van a
    2. are.going.to
    1. llegar
    2. arrive
    1. con
    2. with
    1. como
    2. como
    1. que
    2. que
    1. varios
    2. several
    1. amigos.
    2. friends
    1. ‘They are going to arrive it looks like with several friends.’
    1.  
    1. c.
    1. El
    2. the
    1. Líbano
    2. Lebanon
    1. cuenta
    2. counts
    1. con
    2. with
    1. un
    2. an
    1. ejército
    2. army
    1. de
    2. of
    1. dizque
    2. dizque
    1. 16
    2. 16
    1. mil
    2. thousand
    1. hombres.
    2. men
    1. ‘Lebanon has an army of allegedly 16 thousand men.’
    1.  
    1. d.
    1. El
    2. the
    1. Líbano
    2. Lebanon
    1. cuenta
    2. counts
    1. con
    2. with
    1. un
    2. an
    1. ejército
    2. army
    1. de
    2. of
    1. como
    2. como
    1. que
    2. que
    1. 16
    2. 16
    1. mil
    2. thousand
    1. hombres.
    2. men
    1. ‘Lebanon has an army of it looks like 16 thousand men.’

The flexible distribution of dizque and como que is reminiscent of that of focus-sensitive items in Spanish, such as solo ‘only’. In particular, the distribution of dizque and como que is partially similar to that of solo. Cases with solo are illustrated in (17). Importantly, solo associates with a focused constituent (see Kovacci 1999; Sánchez López 1999; ASALE 2009: 2990–3002 for discussion about Spanish; see also Rooth 1985; Beaver & Clark 2008; Greenberg 2017; 2022; Hirsch 2017; Rochemont 2018). In (17), the constituent can be the full TP completó la primera vuelta (17a), the direct object la primera vuelta (17b) or a PP a las nueve (17c).

    1. (17)
    1. a.
    1. Milagros
    2. Milagros
    1. solo
    2. only
    1. completó
    2. completed
    1. la
    2. the
    1. primera
    2. first
    1. vuelta.
    2. round
    1. ‘Milagros only completed the first round’
    1.  
    1. b.
    1. Milagros
    2. Milagros
    1. completó
    2. completed
    1. solo
    2. only
    1. la
    2. the
    1. primera
    2. first
    1. vuelta.
    2. round
    1. ‘Milagros completed only the first round.’
    1.  
    1. c.
    1. Van
    2. are.going.to
    1. a
    2. arrive
    1. llegar
    2. only
    1. solo
    2. at
    1. a
    2. the
    1. las
    2. nine
    1. nueve.
    1. ‘They are going to arrive only at nine.’

As for cases with solo inside DPs and PPs, solo cannot appear DP-internally marking the NP (18) for focus.13 This contrasts with cases with dizque and como que, where this is possible (15c, d). As for cases inside PPs, solo can appear PP-internally in some cases (19b), but not in others (19a) (see Kayne 1998 for discussion of English only). This distribution contrasts with that of dizque and como que, as illustrated in (16), with the latter appearing inside PPs more broadly (see Riqueros 2013 and Martínez Vera 2020b for discussion of extraction possibilities with PPs in Spanish).

    1. (18)
    1. *Trajo
    2.   brought
    1. un
    2. a
    1. solo
    2. only
    1. compás
    2. compass
    1. marino.
    2. sea
    1.   ‘Juan brought an only sea compass.’
    1. (19)
    1. a.
    1. *Van a
    2.   are.going.to
    1. llegar
    2. arrive
    1. con
    2. with
    1. solo
    2. only
    1. varios
    2. several
    1. amigos.
    2. friends
    1.   ‘They are going to arrive allegedly with several friends.’
    1.  
    1. b.
    1.   El
    2.   the
    1. Líbano
    2. Lebanon
    1. cuenta
    2. counts
    1. con
    2. with
    1. un
    2. an
    1. ejército
    2. army
    1. de
    2. of
    1. solo
    2. only
    1. 16
    2. 16
    1. mil
    2. thousand
    1. hombres.
    2. men
    1.   ‘Lebanon has an army of only 16 thousand men.’(Real Academia Española)

To sum up, we suggest that the (partially) parallel distribution of dizque and como que, and solo (with dizque and como que displaying a more flexible distribution) is explained, because the common property that these elements share is the role they play in connection to focus. This issue will play a key role in our analysis for dizque and como que in section 5. The next section addresses this issue further with regard to presentational focus.

4 Question-answer congruence

The previous section has addressed the flexible distribution of dizque and como que, which is partially reminiscent of elements such as solo, which is focus-sensitive. Based on these parallels, we have suggested that dizque and como que are involved in signaling focus. In this section, we concentrate on presentational focus in particular. This section strengthens this suggestion by taking a closer look at cases of question-answer congruence (Halliday 1967; Hamblin 1973; Kartunnen 1977; Groenendijk & Stokhof 1984; Rooth 1985; 1992; Krifka 1992; Ginzburg 1996; Roberts 1996). We show that dizque and como que are sensitive to previous discourse in that sentences with these elements must be congruent with the previously raised question (intuitively, the question under discussion), i.e., the distribution of dizque and como que is governed by information structure, since they must associate with a constituent that is focused in the discourse context.14 As standardly assumed, a sentence is congruent to a question if the Hamblin set for the question is a subset of the focus alternative set in the answer (for concreteness, we further understand the previous discourse, i.e., the relevant question, as the immediately preceding utterance to the ones under investigation). The discussion has further implications as to how association with the relevant constituent takes place, which builds on the discussion in the previous section. Interestingly, dizque and como que must precede the element that signals the relevant information. Building on the discussion in section 3, we propose that dizque and como que, in addition to being evidentials, are focus markers, which accounts for the question-answer congruence facts in a straightforward way.

We begin the discussion with a What happened? question that asks for Pedro’s whereabouts. The exchange in (20) is congruent in that the answer provides a resolution for the question (if there is no obvious explicit or inferred discourse to which (20A) can be tied, the exchange below would not be felicitous; see (23) below). The elements dizque (20B) and como que (20B’) mark the clause for focus.

    1. (20)
    1. A:
    1. ¿Qué
    2.   what
    1. pasó
    2. happened
    1. (ayer
    2.   yesterday
    1. con
    2. with
    1. Pedro)?
    2. Pedro
    1.   ‘What happened (yesterday with Pedro)?
    1.  
    1. B:
    1.   Dizque
    2.   dizque
    1. (Pedro)
    2.   Pedro
    1. llegó
    2. arrived
    1. de
    2. from
    1. viaje.
    2. trip
    1.   ‘Allegedly, he arrived from his trip.’
    1.  
    1. B’:
    1.   Como
    2.   como
    1. que
    2. que
    1. (Pedro)
    2.   Pedro
    1. llegó
    2. arrived
    1. de
    2. from
    1. viaje.
    2. trip
    1.   ‘It looks like he arrived from his trip.’

Interestingly, dizque and como que must immediately precede the constituent that provides the information that answers the relevant question. If these elements appear in a different position, the answer is infelicitous. This is illustrated below with an object question (21) and an adjunct question (22)—in parentheses we indicate the elements of the answer that would be omitted when providing a short answer. In (21), dizque (21B) and como que (21B’) must precede the object. Otherwise, the answers are infelicitous, as shown in (21B”), where dizque and como que are sentence-initial. (21B, B’) are thus congruent with (21A); the position of dizque and como que is key. Something similar takes place in (22): dizque and como que must precede the time expression that provides the relevant information. If this is not the case, the answers are infelicitous. It is worth noting that it does not seem to be the case that prosody plays a role in connection to signaling the element that provides the relevant information; the prosody of (21B, B’) is no different from declarative sentences in general in the language to the best of our knowledge (see Face 2003 for an overview; see Muñetón Ayala & Dorta 2015 for discussion of declarative sentences in Colombian Spanish in particular); the focused element does not appear to display any prosodic reflex.

    1. (21)
    1. A:
    1. ¿Qué
    2.   what
    1. compró
    2. bought
    1. Pedro?
    2. Pedro
    1.   ‘What did Pedro buy?’
    1.  
    1. B:
    1.   (Pedro
    2.     Pedro
    1. compró)
    2. bought
    1. dizque
    2. dizque
    1. unos
    2. some
    1. zapatos.
    2. shoes
    1.   ‘Pedro bought some alleged shoes.’
    1.  
    1. B’:
    1.   (Pedro
    2.     Pedro
    1. compró)
    2. bought
    1. como
    2. como
    1. que
    2. que
    1. unos
    2. some
    1. zapatos.
    2. shoes
    1.   ‘Pedro bought some kind of shoes.’
    1.  
    1. B”:
    1. #Dizque /
    2.   dizque
    1. Como
    2. como
    1. que
    2. que
    1. Pedro
    2. Pedro
    1. compró
    2. bought
    1. unos
    2. some
    1. zapatos.
    2. shoes
    1. (22)
    1. A:
    1. ¿Cuándo
    2.   when
    1. viene
    2. comes
    1. Pedro?
    2. Pedro
    1.   ‘When will Pedro come?’
    1.  
    1. B:
    1.   (Pedro
    2.     Pedro
    1. viene)
    2. comes
    1. dizque
    2. dizque
    1. a
    2. at
    1. las
    2. the
    1. nueve.
    2. nine
    1.   ‘Pedro comes allegedly at nine.’
    1.  
    1. B’:
    1.   (Pedro
    2.     Pedro
    1. viene)
    2. comes
    1. como
    2. como
    1. que
    2. que
    1. a
    2. at
    1. las
    2. the
    1. nueve.
    2. nine
    1.   ‘Pedro comes it looks like at nine.’
    1.  
    1. B”:
    1. #Dizque /
    2.   dizque
    1. Como
    2. como
    1. que
    2. que
    1. Pedro
    2. Pedro
    1. viene
    2. comes
    1. a
    2. at
    1. las
    2. the
    1. nueve.
    2. nine

Note that these examples are interesting with regard to dizque and como que having to appear adjacent to the constituent that brings in the relevant information. This suggests that the elements under investigation are like those studied in Assmann et al. (2023) in that they show that different focus markers cross-linguistically require that their associate be adjacent to them. This is different from markers such as only in English, where prosody may play a role in that it signals the placement of focus, e.g., in long distance association (Hirsch 2017)—prosody plays no apparent role in the markers studied by Assmann et al. (2023); there may be morphological markers which have to appear next to the focused constituent (this is a strategy than some languages display, which runs in parallel with another strategy where prosody plays a role). Thus, when considering the English sentence Pedro only bought shoes as answer to the question What did Pedro buy?, the direct object in the answer, i.e., shoes, would display a prosodic reflex and long distance association with only is possible. This is not the case for sentences with dizque and como que, as in (21)–(22): they play a crucial role in question-answer congruence by marking the constituent that provides the information that answers the question for focus, which is adjacent to them; prosody does not appear to play a role.15

It is worth noting as well that sentences with dizque and como que must actually be congruent with the previously raised question, i.e., in the absence of such a question (or if the relevant question cannot be reconstructed or inferred), sentences with these elements cannot be felicitously uttered. This is illustrated in (23). Here there is a generic question, which we assume is uttered in a scenario in which it does not raise (or allows the speaker to infer) a particular issue—its purpose is simply to act as a conversation opener. Suppose that the interlocutors have not seen each other in a very long time, and they did not really know much about each other; suppose further that Marta is a common acquaintance, but this is only a piece of shared information among potentially many other pieces of shared information—it is thus not obvious why information about Marta in particular would be communicated in the first place. The contrast in judgment between (23B, B’) vs. (23B”) shows that a simple declarative sentence, i.e., (23B”), can be uttered providing such a piece of information in the exchange (even though there is no obvious prior discourse to which such an utterance contributes). This contrasts with sentences with dizque and como que in that these cannot be uttered. The reason for this is that there is no obvious explicit or inferred discourse to which (23B, B’) contribute, in contrast to (20), where a What happened? question was tied to Pedro’s whereabouts.16

    1. (23)
    1. A:
    1.   Hola,
    2.   hi
    1. ¿qué
    2. what
    1. hay?
    2. is
    1.   ‘Hi, what’s up?’
    1.  
    1. B:
    1. #Dizque
    2.   dizque
    1. Marta
    2. Marta
    1. se
    2. CL
    1. ganó
    2. won
    1. una
    2. a
    1. beca.
    2. fellowship
    1.   ‘Allegedly, Marta got a fellowship.’
    1.  
    1. B’:
    1. #Como
    2.   como
    1. que
    2. que
    1. Marta
    2. Marta
    1. se
    2. CL
    1. ganó
    2. won
    1. una
    2. a
    1. beca.
    2. fellowship
    1.   ‘It looks like Marta got a fellowship.’
    1.  
    1. B”:
    1.   Marta
    2.   Marta
    1. se
    2. CL
    1. ganó
    2. won
    1. una
    2. a
    1. beca.
    2. fellowship
    1.   ‘Marta got a fellowship.’

Note that the evidential contribution must still be present in the cases discussed in this section. Recall in this regard the exchanges in (22), which are repeated in (24). These are only felicitous if B and B’ have the relevant evidence. Thus, (24B) is felicitously uttered if B has reportative evidence, e.g., B heard from someone else that Pedro would come at nine. Similarly, (24B’) is felicitously uttered if B’ has inferential evidence, e.g., B’ knows about Pedro’s whereabouts, in particular, B’ knows that Pedro usually tries to arrive at nine, although he is usually delayed. In the presence of direct, reliable evidence, uttering (24B, B’) is infelicitous. In this regard, suppose that B and B’ are with Pedro, and they are on their way to their destination. It is 8:50 and they are about 10 minutes away from their destination. Somebody calls B and B’ and asks about Pedro’s arrival time. In this scenario, uttering (24B, B’) is infelicitous, which suggests that the evidential contribution is still present in these cases.

    1. (24)
    1. A:
    1. ¿Cuándo
    2.   when
    1. viene
    2. comes
    1. Pedro?
    2. Pedro
    1.   ‘When will Pedro come?’
    1.  
    1. B:
    1. #(Pedro
    2.     Pedro
    1. viene)
    2. comes
    1. dizque
    2. dizque
    1. a
    2. at
    1. las
    2. the
    1. nueve.
    2. nine
    1.   ‘Pedro comes allegedly at nine.’
    1.  
    1. B’:
    1. #(Pedro
    2.     Pedro
    1. viene)
    2. comes
    1. como
    2. como
    1. que
    2. que
    1. a
    2. at
    1. las
    2. the
    1. nueve.
    2. nine
    1.   ‘Pedro comes it looks like at nine.’

Based on the partially overlapping distribution of dizque and como que, and solo (see section 3), in addition to the congruence facts discussed in this section, we propose that dizque and como que are focus markers. They are thus mixed markers that convey evidentiality and focus.

5 Proposal

This section provides an analysis of sentences with dizque and como que. In section 5.1, we discuss the syntactic aspects of constructions with these markers. In section 5.2, we turn to the semantic aspects of our discussion. Our approach provides a unified account for sentences with dizque and como que modifying both clausal and subclauses constituents.

5.1 The syntax of dizque and como que

When dizque and como que modify the clause, we assume that they combine with a proposition denoting expression (i.e., an expression of type st; see section 5.2). Syntactically, we assume that these evidentials are located in C or in some extended version of CP (in, e.g., EvidP or FocP; see Rizzi 1997 and much subsequent work). Demonte & Fernández-Soriano (2022) propose that dizque merges inside TP but is covertly bound by an argument in the Speech Act Phrase (see Speas & Tenny 2003). We depart from this view, suggesting that when dizque and como que modify the clause, they are generated in the higher positions of the sentence where evidential heads are located, i.e., in the CP (see Cruschina & Remberger 2008). In the structure in (25), X marks the evidential’s sister, the TP, for focus; dizque and como que sit higher up in the structure, which we assume is in the CP (Rizzi 1997)—note that dizque and como que would still be taking a proposition denoting complement, the TP here, which is what matters from a semantic point of view (see section 5.2).

    1. (25)
    1. Clausal-scope: [CP dizque / como que [TP … ]X ]

When dizque and como que modify a constituent smaller than the clause, we assume that the evidential moves covertly to the left periphery due to a type mismatch, as shown schematically in (26) (see section 5.2 for additional details).17 In that structure, dizque and como que are initially merged with constituent α, which does not denote a proposition, but, e.g., an individual (type e), a property of eventualities (type vt, with v the type of eventualities), etc. (see section 3). This type mismatch triggers merger with the clausal category TP that has the appropriate semantic type (i.e., st).

    1. (26)
    1. Constituent-scope: [CP dizque / como que [TP … [ t [ α ]X ] … ]]

Our general setup is thus similar to analyses of focus-sensitive items, e.g., only, which is to be expected, since we claim that dizque and como que are mixed markers that make a focus contribution (see Quek & Hirsch 2017 for discussion of approaches along these lines). For instance, Rochemont (2018) proposes a movement analysis of only to an adverbial position above vP. In this sense, only in (27a) would raise to the position that it occupies overtly in (27b) (from Rochemont 2018: 264; adopting the above mentioned convention, X indicates focus). As expected, this movement is locally constrained to the same domain, i.e., the same phase.

    1. (27)
    1. a.
    1. John likes only BillX.
    1.  
    1. b.
    1. John only likes BillX.

That movement is locally constrained to the same domain is unproblematic in many cases with dizque and como que. Thus, the movement illustrated in (26) applies without a problem in the examples in (11)–(13), which are repeated as (28a–30a), since the constituents to which dizque and como que attach can be left-dislocated independently, as seen in (28b)–(30b).

    1. (28)
    1. a.
    1. La
    2. the
    1. lluvia
    2. rain
    1. sigue
    2. keeps
    1. dizque /
    2. dizque
    1. como
    2. como
    1. que
    2. que
    1. cayendo.
    2. falling
    1. ‘The rain keeps allegedly/it looks like falling.’
    1.  
    1. b.
    1. Cayendo,
    2. falling
    1. sigue
    2. keeps
    1. la
    2. the
    1. lluvia.
    2. rain
    1. ‘The rain keeps falling.’
    1. (29)
    1. a.
    1. Va a
    2. it.is.going.to
    1. llover
    2. rain
    1. dizque /
    2. dizque
    1. como
    2. como
    1. que
    2. que
    1. esta
    2. this
    1. noche.
    2. night
    1. ‘It is going to rain allegedly/it looks like tonight.’
    1.  
    1. b.
    1. Esta
    2. this
    1. noche
    2. night
    1. va a
    2. it.is.going.to
    1. llover.
    2. rain
    1. ‘It is going to rain tonight.’
    1. (30)
    1. a.
    1. Van a
    2. are.going.to
    1. llegar
    2. arrive
    1. dizque /
    2. dizque
    1. como
    2. como
    1. que
    2. que
    1. a
    2. at
    1. las
    2. the
    1. nueve.
    2. nine
    1. ‘They are going to arrive allegedly/it looks like at nine.’
    1.  
    1. b.
    1. A
    2. at
    1. las
    2. the
    1. nueve,
    2. nine
    1. van a
    2. are.going.to
    1. llegar.
    2. arrive
    1. ‘They are going to arrive at nine.’

However, as Rochemont (2018: 272) indicates, only cannot raise outside of a prenominal DP position, which means that (31a) does not derive from (31b).

    1. (31)
    1. a.
    1.   Mary only approached the studentX at the party.
    1.  
    1. b.
    1. *Mary approached the only studentX at the party.

Rochemont’s (2018) approach thus raises a question with regard to (26) as to how cases where dizque and como que appear DP- and PP-internally are to be accounted for (see Uriagereka 1988; Corver 1992; Bošković 2005; 2013; Rochemont 2018; Oda 2022 for extensive discussion of extraction out of DPs; see Riqueros 2013 and Martínez Vera 2020b for discussion of extraction possibilities with PPs in Spanish). Recall in this regard examples such as those in (32), which repeat (15c, d). Here, dizque and como que appear inside the DP, between the indefinite determiner un ‘a’ and the NP compás marino ‘sea compass’.

    1. (32)
    1. a.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. un
    2. a
    1. dizque
    2. dizque
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought an alleged sea compass (that ended up being a photo viewer).’
    1.  
    1. b.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. un
    2. a
    1. como
    2. como
    1. que
    2. que
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought a kind of a sea compass (that ended up being a photo viewer).’

In order to account for these cases, we extend Cable’s (2007; 2010) analysis of wh-questions in languages such as Tlingit, Sinhala, Japanese and Korean that appear with an overt Q-particle (sa, da, ka and ka respectively). These particles mark focus in addition to licensing a question, according to Cable, who proposes that Q-particles establish an agreement relation with C (as is well-known, Q-particles and focus elements are tightly knit, as in Cable’s proposal; see Hirsch 2017; Quek & Hirsch 2017; Wu 2022; Branan & Erlewine 2023). In some languages, such as Tlingit and Sinhala, overt wh-movement to Spec, CP follows, whereas in others, such as Japanese and Korean, overt movement does not happen. This difference stems from distinct structural configurations for each language. In wh-movement languages like Tlingit, Q takes an XP containing the wh-word as a complement, as in (33a), while in languages like Japanese and Korean Q is adjoined to XP, as in (33b).18,19

    1. (33)
    1. a.
    1. [QP Q [XP … wh … ]]
    1.  
    1. b.
    1. [XP Q [XP … wh … ]]

Q agrees with C in both types of languages, but in languages with the complement structure in (33a) (Tlingit and Sinhala), QP moves to Spec, CP pied-piping XP. This is illustrated in (34), which schematically exemplifies the fronting of a wh-word in Tlingit in a wh-question after Q-movement (Cable 2007: 86). In (34), there is agreement between CQ and QP, and QP moves overtly to the CP—the QP pied-pipes XP, as made explicit in (34b); (34b) disentangles the relevant components of the QP.

    1. (34)
    1. a.
    1. [CP CQ [IP … QP … ]]
    1.  
    1. b.
    1. [CP [QP [XP … wh-word … ] Q ] [CP CQ [IP … QP … ]]]

In Q-adjoining languages (see (33b)), Q can either overtly move to CP (as in Japanese and Korean) or move covertly, without pied-piping XP. Movement is possible in this case, because Q is adjoined to XP. This is illustrated in (35), which involves Q-movement (which may be overt or covert) in a Q-adjoining language (there is no pied-piping of XP in this case); there is agreement between CQ and Q (Cable 2007: 169).

    1. (35)
    1. a.
    1. [CP [IP … [XP [XP … wh-word … ] Q ] … ] CQ ]
    1.  
    1. b.
    1. [CP [CP [IP … [XP [XP … wh-word … ] Q ] … ] CQ ] Q ]

We propose that dizque and como que have the structure of Q-particles. This follows from the similarity between them and focus-sensitive elements such as solo ‘only’ (see section 3), as well as their role in presentational focus (see section 4). As in Cable’s analysis, como que and dizque move to the higher position in the CP, where they take wide scope over the proposition denoting TP. We assume the initial structure is (33b), which involves covert movement in our analysis—overt movement would yield a structure that is ultimately infelicitous, in particular, in cases with narrow focus, since long distance association in the surface syntax is not available.20 Thus, we render the relevant structure in (33b) as (36).

    1. (36)
    1. [CP Q(dizque) / Q(como que) [TP … [XP Q(dizque) / Q(como que) XP ] … ]]

The fact that dizque and como que can appear between D and NP also argues in favor of structure (36). Cable (2007: 122) proposes that a QP cannot intervene between a functional head and a phrase selected by that functional head. Since languages with the structure in (33a) project a QP, this QP has the potential to intervene between a functional head and its selected lexical complement. In languages with the structure in (33b), on the other hand, Q will not intervene between the functional head and its complement, because Q is adjoined and does not project. Therefore, the Q-particle ka in Korean does not block selection of FocP by Force in (37), which has the relevant Q feature (Cable 2007: 169).

    1. (37)
    1. [ForceP [FocP [FocP … wh-word … ] Q(ka) ] ForceQ ]

Similarly, when dizque or como que appear between D and N they will not block selection of NP by D, because they have the adjoined structure in (38). In this structure, NP is the sister head to D and adjoined dizque and como que do not block selection. This is the relevant structure for cases such as (32), where there is an indefinite D, un ‘a’, and an NP, compás marino ‘sea compass’. Movement of dizque or como que can then proceed as represented in (36).

    1. (38)
    1. [DP D [NP Q(dizque) / Q(como que) NP ]]

5.2 The semantics of dizque and como que

In this section, we provide a semantic account of clausal- and constituent-scope cases with dizque and como que. Our analysis brings three aspects to the foreground. These are the evidence that the speaker must have upon utterance of the scope proposition; a contextual set of alternatives, which differ depending on the evidentials’ scope (i.e., clausal- or constituent-scope); and the status of the alternatives with respect to the scope proposition. We argue that the alternatives are key in connection to the implication of lack of certainty that is present in the cases under consideration. We further suggest that the difference with regard to disbelief, which may be present with dizque but not with como que, is linked to the evidential component in that the difference between reportative and inferential evidence has a consequence for the utterer’s stance with regard to the scope proposition. In what follows, we explain these components; these are brought together in (42) and (43), where the denotations of dizque and como que are given.

As for the evidence that the speaker has, following previous literature, we assume that dizque and como que encode hearsay and inferential evidence respectively (see Murray 2010; 2017 for an overview of evidentials encoding these kinds of evidence). As discussed in section 2, dizque is compatible with second- and third-hand evidence, i.e., it is compatible with a wide array of possibilities in connection to reports (e.g., the source may or may not be known, etc.). Here we capture the evidential contribution of dizque by means of the relation Rep, which takes speaker sp and proposition p as arguments, Rep(sp, p), and holds iff sp has reportative evidence for p, where the reportative evidence is understood in a general, underspecified manner (see Demonte & Fernández-Soriano 2022; Martínez Vera 2023 for recent discussion). In addition, we assume that the scope proposition is presented (in an exchange to e.g., address the question under discussion; see Simons et al. 2010; see section 4); this follows Faller (2002; 2019). Importantly, as Faller (2019) notes, presentation basically removes the inference of sincerity that the speaker believes the scope proposition (Vanderveken 1990). To make this explicit, we adopt a Present relation that takes speaker sp and proposition p, Present(sp, p), and holds iff sp presents p, which is to say that p is put out there (to address the question under discussion in an exchange), and crucially means that that sp remains silent as to whether she believes p.21

This captures that, while the speaker may present some proposition based on reportative evidence, she may not believe such a proposition or may remain undecided with respect to it (see AnderBois 2014; Murray 2017; Faller 2019; Pancheva & Rudin 2019 for discussion of reportative evidentials, in particular, with regard to the absence of necessary commitment to the truth of the scope proposition). Adopting this position accounts for the felicitous continuations in (6a) (the example is repeated in (39)), which indicates that the speaker does not believe the scope proposition or is uncertain in this regard.

    1. (39)
    1. Dizque
    2. dizque
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche,
    2. night
    1. pero
    2. but
    1. yo
    2. I
    1. no
    2. no
    1. lo
    2. CL
    1. creo /
    2. believe
    1. no
    2. no
    1. estoy
    2. am
    1. seguro
    2. sure
    1. de
    2. of
    1. si
    2. if
    1. así
    2. like.that
    1. será.
    2. will.be
    1. ‘Allegedly, it is going to rain tonight, but I don’t believe it/I’m not sure whether it’ll actually rain.’

In turn, como que differs from dizque in that it requires inferential evidence. As discussed in section 2, como que is compatible with inferences that are drawn from relevant cues (e.g., results), as well as common knowledge—the speaker thus makes a conjecture with regard to a state of affairs (Faller 2002; 2023). Here we capture the evidential contribution of como que by means of the relation Inf, which takes speaker sp and proposition p as arguments, Inf(sp, p), and holds iff sp has inferential evidence for p, which intends to capture different possibilities in terms of inferences that can be drawn based on relevant evidence (see Grajales 2017). In this case, the evidence relies on the speaker’s judgment, i.e., the speaker makes an inference or draws a conclusion based on some relevant evidence available to her. Following Faller (2002; 2023) (see also Izvorski 1997; Matthewson et al. 2007), we assume that in the presence of an inferential evidential, the speaker considers that the scope proposition is epistemically possible, which would be similar to English might.22 We incorporate this in connection to assertions (Faller 2002), in particular, sentences with como que involve assertions of possibly p (in an exchange to address the question under discussion). Importantly, asserting a proposition involves the sincerity conditions that the speaker believes such a proposition (Faller 2002), in this case, that she believes that such a proposition is possible. We capture this as Assert(sp, ◇p), which holds iff sp asserts that p is possible, which importantly means that sp believes that p is possible.

Since the speaker considers that the scope proposition is possible, the facts in (6b) ((6b) is repeated in (40)) are accounted for. In this case, the speaker may indicate that she is uncertain with respect to the scope proposition, but she cannot say that she does not believe it (as she deems such a proposition possible).

    1. (40)
    1. Como
    2. como
    1. que
    2. que
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche,
    2. night
    1. pero
    2. but
    1. #yo
    2.   I
    1. no
    2. no
    1. lo
    2. CL
    1. creo /
    2. believe
    1. no
    2. no
    1. estoy
    2. am
    1. seguro
    2. sure
    1. de
    2. of
    1. si
    2. if
    1. así
    2. like.that
    1. será.
    2. will.be
    1. ‘It looks like it is going to rain tonight, but I don’t believe it/I’m not sure whether it’ll actually rain.’

Turning to the consideration of alternatives, as discussed in sections 3–4, dizque and como que are similar to focus-sensitive elements (Demonte & Fernández Soriano 2022) such as solo ‘only’ (Rooth 1985; 1992; 2016; Greenberg 2017; 2022; Hirsch 2017). Recall the structures in (25)–(26), repeated in (41). As discussed in section 5.1, we assume that dizque and como que mark the element in their scope (i.e., its sister, which is adjacent to it; see Assmann et al. 2023), which we represent by means of X. In (41a), these elements are base-generated in the relevant position in the left periphery and the whole clause is marked (recall that we assume that TP is a proposition denoting element), whereas, in (41b), they are base-generated taking constituent-scope (and thus mark the relevant constituent) and will move covertly to the left periphery due to a type mismatch. We further assume that, in (41b), the evidentials leave a vacuous trace upon movement (Wilkinson 1996; Nakanishi 2012; see also Hirsch 2017).

    1. (41)
    1. a.
    1. Clausal-scope: [CP dizque / como que [TP … ]X ]
    1.  
    1. b.
    1. Constituent-scope: [CP dizque / como que [TP … [ t [ α ]X ] … ]]

The contextual set of alternatives (i.e., the focus value of the sentences under consideration) that will be considered is determined by means of what element is marked by dizque and como que (a similar approach has been applied to evidentiality by Martínez Vera 2024; we do not explicitly represent context for simplicity; see Rooth 1985; 1992). Here we illustrate this with two cases, namely, a case where the clause is marked (e.g., (1)), and a case where the marked constituent denotes an entity (e.g., (2)). If the whole clause is marked (e.g., in (1)), then the set of alternatives to be considered includes alternative propositions (the denotation of the relevant clause is also in this set). Thus, if the relevant clause that is marked by dizque or como que denotes the scope proposition that it is going to rain tonight (i.e., this is the clause’s ordinary value), then Alt(⟦[TP It is going to rain tonight ]X⟧) = {q | q ∈ Dst} (English is used for readability). If, on the other hand, dizque and como que mark a (subclausal) constituent (e.g., in (2)), then the alternatives are propositions that vary in that element. For instance, if dizque and como que mark an expression that denotes the individual the sea compass in the relevant clause, then the alternatives vary with regard to this individual, e.g., Alt(⟦[TP Juan brought [ the sea compass ]X ]⟧) = {λw[Juan brought x in w] | x ∈ De}.

What remains to be addressed is the status of the alternatives with respect to the scope proposition, which in our approach is fundamental in connection to the implication of lack of certainty. Adopting a suggestion made to us by an anonymous reviewer, we propose that there is an alternative q different from scope proposition p such that it is possible that only q is true.23 We thus get that p is presented in the case of dizque or that ◇p is asserted in the case of como que, and that, simultaneously, q may possibly be the only true proposition among alternatives (we represent the latter as ◇[Only(q)]). If this is the case, then, it would be possible that q by itself addresses (e.g., provides a complete answer to) the question under discussion. If Only(q) held, then p would be false, so ◇[Only(q)] would entail that it is possible that ¬p, which is equivalent to not believing p. By means of illustration, consider the sentences in (2), i.e., Juan trajo dizque/como que un compás marino ‘Juan brought a kind of a sea compass’. Here ◇[Only(q)] would say that it is possible that Juan brought something other than a sea compass and nothing else, which leads to the implication that it is not certain that Juan brought a sea compass.24

The denotations of dizque and como que appear in (42) and (43) respectively. Dizque applies to a proposition; the meaning conveyed is that the speaker presents that proposition and has reportative evidence for it, and it is possible that some alternative to that proposition is the only true one. Como que applies to a proposition; the meaning conveyed is that the speaker asserts that that proposition is possible and has inferential evidence for it, and it is possible that some alternative to that proposition is the only true one.25

    1. (42)
    1. ⟦dizque⟧(p) = Present(sp, p) & Rep(sp, p) & ∃q ∈ Alt(p)[q ≠ p & ◇[Only(q)]]
    1. (43)
    1. ⟦como que⟧(p) = Assert(sp, ◇p) & Inf(sp, p) & ∃q ∈ Alt(p)[q ≠ p & ◇[Only(q)]]

To illustrate our proposal, consider the following examples, where (44) (=(1a)) is a case of clausal-scope and (45) (=(2b)) is a case of constituent-scope. (44) is an example with dizque, where the speaker has reportative evidence for the scope proposition. In this case, the alternatives that are considered are whole propositions (44c, d), i.e., this is a case of clausal-scope (in this example, the scope proposition is marked with X (44b); we further include a short preamble based on the discussion about congruence in section 4). For illustration, suppose that the question under discussion is whether it is going to rain tonight, so the alternatives are the scope proposition and its negation. The implication of lack of certainty is to be understood in terms of such an alternative being the only true one. In this example, the speaker may remain agnostic with regard to the scope proposition or may even believe that it is false.

    1. (44)
    1. a.
    1. (Te
    2. (CL
    1. cuento
    2. tell
    1. lo
    2. the
    1. relevante:)
    2. relevant
    1. dizque
    2. dizque
    1. va a
    2. is.going.to
    1. llover
    2. rain
    1. esta
    2. this
    1. noche.
    2. night
    1. ‘(I’ll tell you what is relevant:) allegedly, it is going to rain tonight.’
    1.  
    1. b.
    1. [CP dizque [TP it is going to rain tonight ]X ]
    1.  
    1. c.
    1. ⟦(44a)⟧ = Present(sp, λw[it is going to rain tonight in w]) &
    2.                        Rep(sp, λw[it is going to rain tonight in w]) &
    3.                                      ∃q ∈ Alt(⟦[TP it is going to rain tonight ]X⟧)
    4.                                               [q ≠ λw[it is going to rain tonight in w]) & ◇[Only(q)]]
    1.  
    1. d.
    1. Alt(⟦[TP it is going to rain tonight ]X⟧) = {q | q ∈ Dst}

The example in (45) illustrates a case of constituent-scope. This is a case with como que, where the speaker has inferential evidence for the scope proposition. The focused constituent is a DP—this phrase is thus marked with X. The alternatives that are considered are propositions that vary with regard to entities (45d) (we assume that the DP denotes an entity of type e for ease of exposition). The implication of lack of certainty arises, because it is possible that an alternative other than the scope proposition is the only true one—in this case, for instance, an alternative that indicates that a photo viewer was brought would be the only possible alternative that is true. In contrast to the previous example with dizque, in this case the speaker cannot remain agnostic with regard to the scope proposition; she believes that such a proposition is actually possible—as a result, disbelief with respect to the scope proposition will not arise.

    1. (45)
    1. a.
    1. Juan
    2. Juan
    1. trajo
    2. brought
    1. como
    2. como
    1. que
    2. que
    1. un
    2. a
    1. compás
    2. compass
    1. marino
    2. sea
    1. (que
    2.   that
    1. resultó
    2. resulted
    1. ser
    2. be
    1. un
    2. a
    1. visor
    2. viewer
    1. de
    2. of
    1. fotos).
    2. photos
    1. ‘Juan brought a kind of a sea compass (that ended up being a photo viewer).’
    1.  
    1. b.
    1. [CP como que [TP Juan brought t [ a sea compass ]X ]
    1.  
    1. c.
    1. ⟦(45a)⟧ = Assert(sp, ◇λw[Juan brought a sea compass in w]) &
    2.                       Inf(sp, λw[Juan brought a sea compass in w]) &
    3.                                        ∃q ∈ Alt(⟦[TP Juan brought [a sea compass]X ]⟧)
    4.                                                [q ≠ λw[Juan brought a sea compass in w]) & ◇[Only(q)]]
    1.  
    1. d.
    1. Alt(⟦[TP Juan brought [ a sea compass ]X ]⟧) = {λw[Juan brought x in w] | x ∈ De}

The discussion in this and the previous section has thus proposed a unified analysis of dizque and como que for cases with clausal- and constituent-scope, thus accounting for the so-called labeling function (Travis 2006). A key contrast between dizque and como que, in our approach, lies in the nature of the evidence associated with each marker, namely, reportative evidence for dizque and inferential evidence for como que. Furthermore, while in both cases an implication of lack of certainty does arise, we have suggested that there is difference with regard to the possibility of there being disbelief: what we find is that, in the case of sentences with dizque, the speaker may believe that the scope proposition is false; in contrast, in the case of sentences with como que, the speaker deems that the scope proposition is possible.

6 Conclusion

In this paper, we have proposed a unified account of matrix declarative sentences with dizque and como que in Colombian Spanish. Our focus has been on the fact that these evidentials can modify clausal and subclausal constituents (Travis 2006). In addition to providing discussion as to what evidence is required in the presence of dizque and como que, our approach has likened these evidentials to focus-sensitive items, such as Spanish solo ‘only’. In this regard, we have further discussed question-answer congruence in connection to these markers, and have shown that sentences with them must be congruent with prior discourse. We provided a syntactic analysis that accounted for the distribution of dizque and como que, including the consideration of (apparent) locality violations. We further provided a semantic analysis of these elements that made explicit their evidential meaning, as well as the role that focus played. The latter was deemed key with regard to calculating the implication of lack of certainty that is tied to sentences with these evidentials.

Some issues are left open for further investigation. Here we mention three. The first one concerns seldom cases where dizque and como que appear in examples where the speaker appears to have direct evidence; what is key is that she distances herself from what is communicated (see footnote 4). The main question here is how to capture the notion of distance and how this is reconciled with the fact that, in most cases, dizque requires reportative evidence and como que requires inferential evidence. The second one concerns discussion of the different layers of meaning that are involved in the denotation of dizque and como que. We have only briefly discussed the issue of question-answer congruence with regard to the scope proposition, while the status of the evidential contribution and alternatives have been mentioned only in passing (see footnote 25). The third issue concerns the analysis of cases with multiple focus-sensitive elements. While here we have concentrated on sentences with one element only, cases such as (46) are possible. Examples like this one raise syntactic and semantic questions with regard to the interaction of the relevant elements that are yet to be addressed in detail.

    1. (46)
    1. Dizque
    2. dizque
    1. José
    2. José
    1. compró
    2. bought
    1. solo
    2. only
    1. una
    2. a
    1. lavadora.
    2. washing machine
    1. ‘Allegedly, José bought only a washing machine.’

Acknowledgements

We would like to thank Agata Renans, the editor of the paper, and three anonymous reviewers for their feedback. We would also like to thank Raúl Bendezú, Roger Schwarzschild, Michelle Sheehan, Tania Torres Oyarce and Malte Zimmermann for discussion.

Competing interests

The authors have no competing interests to declare.

Notes

  1. The data that are reported in this paper are the result of a combination of published sources, with an emphasis on Grajales (2017), who reports data from semi-directed interviews from the PRESEEA-Medellín corpus (see Grajales 2017 for further discussion with regard to methodology); the CREA corpus also documents various cases. We have also found examples with these markers in social media. Additional grammaticality judgements and examples come from one of the co-authors, originally from Bogotá, Colombia. The tests that are reported throughout follow Martínez Vera’s (2023) elicitation practices, as well as his examples. Expressions with dizque and como que are usually used in an informal, oral register, but they are not socially stigmatized or associated with rural speech. [^]
  2. While English translations do not always convey the evidential or inferential meaning of dizque and como que, they are very much present, so the reader should not conclude that these evidentials are simply hedges based on the translations of the examples. [^]
  3. As indicated, we focus on Colombian Spanish. An anonymous reviewer notes that como que is possibly attested as well in Iberian Spanish in cases in which it modifies the whole clause, but not in cases in which it modifies subclausal constituents. If true, the discussion that follows would apply to that variety as well (in particular, for cases where this element modifies the clause). [^]
  4. There are a few instances of direct evidence where sentences with dizque appear to be felicitous, as in (ia). Key here is that the speaker establishes distance with regard to what is uttered. Thus, (ia) is felicitously uttered if the speaker distances herself from what she did, e.g., the speaker went to the gym but did not do the things that are usually done there (such as working out); instead, she just hung out. Interestingly, como que is also felicitous in cases like this one (ib). These examples are set aside in this paper.
      1. (i)
      1. a.
      1. Dizque
      2. dizque
      1. fui
      2. went
      1. al
      2. to.the
      1. gimnasio
      2. gym
      1. esta
      2. this
      1. mañana.
      2. morning
      1. ‘Allegedly, I went to the gym this morning.’
      1.  
      1. b.
      1. Como
      2. como
      1. que
      2. que
      1. fui
      2. went
      1. al
      2. to.the
      1. gimnasio
      2. gym
      1. esta
      2. this
      1. mañana.
      2. morning
      1. ‘It looks like I went to the gym this morning.’
    [^]
  5. We do not discuss quotation and (free) indirect speech in this paper (as indicated, we only focus on matrix sentences) (see Demonte & Fernández-Soriano 2022; Martínez Vera 2023 for recent discussion). To the best of our knowledge, these markers cannot be used to quote other people’s speech in the varieties under consideration. [^]
  6. It is worth pointing out that the counterpart of (5) without dizque and como que is felicitous, which indicates that the test does in fact target whether there is (lack of) certainty. We would like to thank an anonymous reviewer for bringing this to our attention.
      1. (i)
      1. Va a
      2. is.going.to
      1. llover
      2. rain
      1. esta
      2. this
      1. noche;
      2. night
      1. (yo)
      2.   I
      1. no
      2. no
      1. tengo
      2. have
      1. dudas
      2. doubts
      1. al
      2. to.the
      1. respecto.
      2. respect
      1. ‘It is going to rain tonight; I don’t have any doubts about it.’
    [^]
  7. This discussion can be further expanded to the consideration of public commitments (Farkas & Bruce 2010; see Murray 2017; Faller 2019; 2023; Bhadra 2020; Martínez Vera 2024 for discussion about evidentials). This property may be tested by means of a continuation by which the speaker makes a(n overt) statement that questions the truth of the scope proposition. In the presence of dizque (ia), the continuation is possible; with como que (ib), it is not—the speaker must be committed to (the possibility of) the truth of the scope proposition here.
      1. (i)
      1. a.
      1. Dizque
      2. dizque
      1. va a
      2. is.going.to
      1. llover
      2. rain
      1. esta
      2. this
      1. noche,
      2. night
      1. pero
      2. but
      1. no
      2. no
      1. me
      2. CL
      1. parece
      2. seem
      1. que
      2. that
      1. sea
      2. be
      1. verdad.
      2. truth
      1. ‘Allegedly, it is going to rain tonight, but it doesn’t seem true (to me).’
      1.  
      1. b.
      1. Como
      2. como
      1. que
      2. que
      1. va a
      2. is.going.to
      1. llover
      2. rain
      1. esta
      2. this
      1. noche,
      2. night
      1. pero
      2. but
      1. #no
      2.   no
      1. me
      2. CL
      1. parece
      2. seem
      1. que
      2. that
      1. sea
      2. be
      1. verdad.
      2. truth
      1. ‘It looks like it is going to rain tonight, but it doesn’t seem true (to me).’
    A detailed discussion of this property should ultimately be supplemented with detailed discussion of the illocutionary force of sentences with dizque and como que. There is extensive work on reportative evidentials and how they need not entail the commitment to the truth of the scope proposition (see Faller 2002; 2019; Anderbois 2014; among others). This is different from other evidentials, such as inferentials, where such a dissociation has not been documented. While we address these issues to an extent in section 5.2, we leave a more detailed discussion for future research. [^]
  8. An anonymous reviewer points out to us that (7b) appears to not require commitment to the possibility of the scope proposition. They suggest that (7b) can be accommodated if “the modal claim in (7b) has past temporal perspective (see[,] e.g., Rullmann & Matthewson 2018 on epistemic modals with past temporal perspective). In this case, the reference time for the modal claim would be in the past and so the speaker would be required to be committed to the possibility of the truth of the prejacent at that past time but wouldn’t need to be committed to the possibility the prejacent at the utterance time.” We would like to thank the reviewer for this suggestion. We leave the issue of temporal anchoring for further research. [^]
  9. A reviewer suggests that the markers may modify a vP or VP rather than TP. For examples like (9), this would involve the verb not raising to T, contrary to what is generally assumed for Spanish (see Suñer 1994; Ordóñez 1997; Zagona 2002; Camacho & Sánchez 2017). Demonte and Fernández-Soriano (2022) explicitly propose that dizque occupies the same position where auxiliaries are generated inside TP. [^]
  10. Thus, dizque and como que may also modify the subject (9) as well, and the object (see (21)). [^]
  11. There is one exception with PPs, as shown in (i), where the preposition a is present. It seems that the presence of dizque and como que breaks the syntactic-prosodic requirements of the preposition a, namely, that a forms a prosodic unit with the DP complement; this requires further research.
      1. (i)
      1. a.
      1. ??Van a
      2.    are.going.to
      1. llegar
      2. arrive
      1. a
      2. at
      1. dizque
      2. dizque
      1. las
      2. the
      1. nueve.
      2. nine
      1.    ‘They are going to arrive allegedly at nine.’
      1.  
      1. b.
      1. ??Van a
      2.    are.going.to
      1. llegar
      2. arrive
      1. a
      2. at
      1. como
      2. como
      1. que
      2. que
      1. las
      2. the
      1. nueve.
      2. nine
      1.    ‘They are going to arrive it looks like at nine.’
    [^]
  12. To the best of our knowledge, the evidential contribution is still of the whole proposition (see section 2). [^]
  13. There are instances in which it is possible to have solo DP-internally. This is, however, a different kind of solo, which actually agrees with the noun, suggesting that it is an adjective; this item means ‘mere, unique’. See Sánchez (1995) and ASALE (2009) for further discussion; see Bernstein (1993) for related discussion for Italian; see Coppock & Beaver (2014) for an analysis of English mere (and other exclusives).
      1. (i)
      1. a.
      1. Me
      2. CL
      1. sorprende
      2. surprises
      1. la
      2. the
      1. sola
      2. mere
      1. pregunta.
      2. question
      1. ‘The mere question surprises me.’
      1.  
      1. b.
      1. Se
      2. CL
      1. presentó
      2. arrived
      1. con
      2. with
      1. la
      2. the
      1. sola
      2. mere
      1. compañía
      2. company
      1. de
      2. of
      1. un
      2. a
      1. escolta.
      2. bodyguard
      1. ‘S/he arrived with only the company of a bodyguard.’(ASALE 2009: 958)
    [^]
  14. For simplicity, we only discuss cases involving complete answers. While partial answers are possible as long as they are relevant for the question under discussion, they are set aside in this paper. [^]
  15. The element solo in Spanish (see section 3) appears to behave like only in this regard in that long distance association can take place. This is exemplified in (i), where zapatos ‘shoes’ is prosodically prominent. This is different from cases with dizque and como que, as shown in the main text, where prosodic factors do not appear to play a role and long distance association does not take place (see Assmann et al. 2023). A detailed discussion and analysis of the contrasts among different focus-sensitive items in Spanish is left for future research.
      1. (i)
      1. A:
      1. ¿Qué
      2.   what
      1. compró
      2. bought
      1. Pedro?
      2. Pedro
      1.   ‘What did Pedro buy?’
      1.  
      1. B:
      1.   Pedro
      2.   Pedro
      1. solo
      2. only
      1. compró
      2. bought
      1. zapatos.
      2. shoes
      1.   ‘Pedro only bought shoes.’
    [^]
  16. In the absence of (23A), it is also infelicitous to utter (23B, B’), i.e., sentences with dizque and como que cannot be felicitously uttered out of the blue. In contrast, (23B”) is in principle possible. [^]
  17. Following Coppock & Beaver (2014), an analysis for dizque and como que in which they stay in situ is conceivable. We have not taken this route, because we think that it makes it perhaps more transparent, within an approach that highlights the syntax-semantics interface, how these elements make a dual contribution, with the evidential meaning always taking scope over the whole proposition (in contrast to the focus contribution, where a subclausal constituent may be marked). See section 5.2 for the semantic details of our analysis. We would like to thank an anonymous reviewer for discussion in this regard. [^]
  18. An anonymous reviewer suggests that it may be possible to maintain an agreement analysis in which there is a silent operator with which dizque and como que agree. In the cases in which dizque and como que surface in a low position, it would be indicated that these markers can only spell out the low position, in contrast to, e.g., only in Hirsch’s (2017) analysis, where such an element can optionally spell out both high and low positions. We leave a detailed exploration of such an alternative, as well as the comparison of the analyses, for a future occasion. [^]
  19. A separate question is whether the proposed analysis violates locality restrictions on extraction from subjects, for example, as a reviewer asks (Ross 1967; Huang 1982). Note that in-situ wh-questions are much more acceptable in those contexts (i)—wh-in-situ in cases like this shows that there is no violation of the CED, which suggests that no violation should arise for other cases as well, such as those with dizque and como que. Reglero & Ticio (2013) have proposed an analysis of in-situ wh-questions based on Cable’s (2010) proposal.
      1. (i)
      1. ¿[Que
      2.     that
      1. el
      2. the
      1. gato
      2. cat
      1. arañe
      2. scratches
      1. a quién]
      2. who
      1. te
      2. CL
      1. molesta
      2. bothers
      1. tanto?
      2. so much
    [^]
  20. We do not maintain that all instances of dizque and como que originate in a low position (see sections 2 and 3 for discussion of cases where these elements appear in, e.g., sentence-initial position), contrary to Rochemont’s (2018) account for only, although nothing hinges on this. We leave this and other differences in the distribution of dizque and como que, and solo ‘only’ for future research (see sections 3–4). [^]
  21. Our approach is static and has not adopted any machinery that makes reference to a table (Farkas & Bruce 2010), or even the Common Ground (Stalnaker 1978; 2002). Our proposal to have a Present relation tries to capture the fact that the scope proposition is put out there to address the question under discussion, but it is not asserted (Faller 2002; 2019). See Bhadra (2020) and Martínez Vera (2024) for discussion of compositionality in connection to illocutionary force and evidentiality. [^]
  22. This can be captured in different ways. Here we mention, as an explicit example, Faller’s (2023) recent discussion which concentrates on inferential evidentials (this proposal is developed in a Kratzerian approach to modality). In particular, the scope proposition is supported by reasoning (of the speaker) such that there is a world w’ in the set of epistemically accessible worlds ⋂f(w) such that p is true in w’ (f is a contextual, non-empty epistemic modal base and w is the evaluation world). [^]
  23. See Greenberg (2017; 2022) for discussion as to whether a scalar component may be needed in connection to focus-sensitive items (see also Beck 1997; Beaver & Clark 2008; Faller 2012; Coppock & Beaver 2014). See Yalcin (2007), Villalta (2008) and Anand & Hacquard (2013) for discussion as to how to capture the meaning of doubt in Spanish and cross-linguistically. [^]
  24. The account discussed above is compatible with cases in which the scope proposition and an alternative are mutually exclusive. This is the case if, e.g., the alternatives are the scope proposition and its negation (e.g., if the alternatives are based on a question such as whether it is raining for an example such as (1)), but also for cases in which, in principle, the alternatives are not mutually exclusive, as in a case such as the one involving a sea compass (2). Naturally, as in the setup for (2), alternatives are contextually mutually exclusive, and the proposal discussed in the main text can capture such a case. However, mutual exclusivity of alternatives is not necessary. Consider (i) below. Suppose that the alternatives vary in the object being brought, a sea compass or a photo viewer (logically, these alternatives are not mutually exclusive, in contrast to, e.g., the alternatives of a polar question, as in the rain case mentioned above). It is actually possible to follow (i) up with something like En efecto, sí era un compás marino y además trajo un visor de fotos ‘In fact, it actually was a sea compass and he also brought a photo viewer,’ where the possibility of the scope proposition is strengthened and the other alternative is overtly stated.
      1. (i)
      1. Juan
      2. Juan
      1. trajo
      2. brought
      1. como
      2. como
      1. que
      2. que
      1. un
      2. a
      1. compás
      2. compass
      1. marino.
      2. sea
      1. ‘Juan brought a kind of a sea compass.’
    The current proposal can accommodate these facts, since what is required is the mere possibility that an alternative distinct from the scope proposition be the only true one. We would like to thank an anonymous reviewer for making very useful and insightful suggestions with regard to how to capture the implication of lack of certainty. [^]
  25. While the denotations do not include a layered approach (e.g., we have not addressed the at-issue vs. not-at-issue distinction; see, e.g., Murray 2017 for an overview), here we provide some insights in this regard based on a suggestion by an anonymous reviewer. There is an illocutionary meaning involving the relations Present (in the case of dizque) and Assert (in the case of como que) which indicates what is at-issue, namely, proposition p and ◇p, respectively. The latter has been explicitly addressed in connection to answerhood (Koev 2018; Korotkova 2020) in section 4. While we have not addressed the at-issueness status of the evidential contributions or the alternatives, following Martínez Vera (2023), we suggest that these meanings are not-at-issue. [^]

References

Aikhenvald, Alexandra. 2004. Evidentiality. New York: Oxford University Press. DOI:  http://doi.org/10.1093/oso/9780199263882.001.0001

Aikhenvald, Alexandra. 2006. Evidentiality in Grammar. In Brown, Keith (ed.), Encyclopedia of language and linguistics, 320–325. Oxford: Elsevier. DOI:  http://doi.org/10.1016/B0-08-044854-2/00252-2

Aikhenvald, Alexandra. 2018. The Oxford handbook of evidentiality. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/oxfordhb/9780198759515.001.0001

Alcázar, Asier. 2014. On the grammaticalization of dizque. In Arias, Andrés Enrique & Gutiérrez, Manuel J. & Landa, Alazne & Ocampo, Francisco (eds.), Perspectives in the study of Spanish Language Variation, 20–42. Santiago de Compostela: Universidad de Santiago de Compostela.

Alcázar, Asier. 2018. Dizque and other emergent evidential forms in Romance languages. In Aikhenvald, Alexandra (ed.), The Oxford handbook of evidentiality, 725–740. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/oxfordhb/9780198759515.013.35

Anand, Pranav & Hacquard, Valentine. 2013. Epistemics and attitudes. Semantics and Pragmatics 6(8). 1–59. DOI:  http://doi.org/10.3765/sp.6.8

AnderBois, Scott. 2014. On the exceptional status of reportative evidentials. Semantics and Linguistic Theory 24. 234–254. DOI:  http://doi.org/10.3765/salt.v24i0.2424

Asociación de Academias de la Lengua Española (ASALE). 2009. Nueva gramática de la lengua española. Vol. II Sintaxis. Madrid: Espasa-Calpe.

Assmann, Muriel & Büring, Daniel & Jordanoska, Izabela & Prüller, Max. 2023. Towards a theory of morphosyntactic focus marking. Natural Language & Linguistic Theory 41. 1349–1396. DOI:  http://doi.org/10.1007/s11049-023-09567-4

Babel, Anna. 2009. Dizque, evidentiality, and stance in Valley Spanish. Language in Society 38. 487–511. DOI:  http://doi.org/10.1017/S0047404509990236

Beaver, David & Clark, Brady. 2008. Sense and sensitivity. Oxford: Blackwell. DOI:  http://doi.org/10.1002/9781444304176

Beck, Sigrid. 1997. On the semantics of comparative conditionals. Linguistics and Philosophy 26. 511–545.

Bernstein, Judy. 1993. Topics in the syntax of nominal constructions across Romance. The City University of New York dissertation.

Bhadra, Diti. 2020. The semantics of evidentials in questions. Journal of Semantics 37. 367–423. DOI:  http://doi.org/10.1093/jos/ffaa003

Bošković, Željko. 2005. On the locality of left branch extraction and the structure of NP. Studia Linguistica 50. 1–45. DOI:  http://doi.org/10.1111/j.1467-9582.2005.00118.x

Bošković, Željko. 2013. Phases beyond clauses. In Schürcks, Lilia & Giannakidou, Anastasia & Etxeberria, Urtzi (eds.), The nominal structure in Slavic and beyond, 75–128. Berlin: Mouton de Gruyter. DOI:  http://doi.org/10.1515/9781614512790.75

Branan, Kenyon & Erlewine, Michael Y. 2023. Anti-pied-piping. Language 99(3). 603–653. DOI:  http://doi.org/10.1353/lan.2023.a907013

Cable, Seth. 2007. The Grammar of Q: Q-Particles and the Nature of Wh-Fronting, As Revealed by the Wh-Questions of Tlingit. Massachusetts Institute of Technology dissertation.

Cable, Seth. 2010. The grammar of Q: Q-particles, wh-movement and pied-piping. Oxford: Oxford University Press.

Camacho, José. 2011. La estructura de la secuencia como que. In Escandell, María Victoria & Leonetti, Manuel & Sánchez López, Cristina (eds.), 60 problemas de gramática. Madrid: Akal.

Camacho, José & Sanchez, Liliana. 2017. Does the verb raise to T in Spanish? In Fernández-Soriano, Olga & Castroviejo, Elena & Pérez-Jiménez, Isabel (eds.), Boundaries, phases, and interfaces, 48–61. Amsterdam: Benjamins.

Coppock, Elizabeth & Beaver, David. 2014. Principles of the exclusive muddle. Journal of Semantics 31. 371–432. DOI:  http://doi.org/10.1093/jos/fft007

Corver, Norbert. 1992. On deriving left branch extraction asymmetries: A case study in paramentric syntax. In Broderick, Kimberly (ed.), Proceedings of the 22nd Conference of the North East Linguistic Society, 67–84. Amherst: GLSA.

Cruschina, Silvio & Remberger, Eva-Maria. 2008. Hearsay and reported speech: Evidentiality in Romance. Rivista di Grammatica Generativa 33. 95–116.

De la Mora, Juliana & Maldonado, Ricardo. 2015. Dizque: Epistemics blurring evidentials in Mexican Spanish. Journal of Pragmatics 85. 168–180. DOI:  http://doi.org/10.1016/j.pragma.2015.03.019

Demonte, Violeta & Fernández-Soriano, Olga. 2013. Evidentials dizque and que in Spanish: grammaticalization parameters, and the (fine) structure of Comp. Linguística: Revista de Estudos linguísticos da Universidad do Porto 8. 211–234.

Demonte, Violeta & Fernández-Soriano, Olga. 2020. Dizque. Un evidencial reportativo modal en la periferia izquierda oracional. In Hernández Palacios, Esther & Martín-Butragueño, Pedro (eds.), Elementos de variación lingüística. La palabra, 185–218. México, D.F.: CSIC – El Colegio de México.

Demonte, Violeta & Fernández-Soriano, Olga. 2022. A multidimensional analysis of the Spanish reportative epistemic evidential dizque. Lingua 266. 103168. DOI:  http://doi.org/10.1016/j.lingua.2021.103168

Dendale, Patrick & Tasmowski, Liliane. 2001. Introduction: Evidentiality and related notions. Journal of Pragmatics 33. 339–348. DOI:  http://doi.org/10.1016/S0378-2166(00)00005-9

Face, Timothy L. 2003. Intonation in Spanish declaratives: Differences between lab speech and spontaneous speech. Catalan Journal of Linguistics 2. 115–131. DOI:  http://doi.org/10.5565/rev/catjl.46

Faller, Martina. 2002. Semantics and pragmatics of evidentials in Cuzco Quechua. Stanford University dissertation.

Faller, Martina. 2012. Evidential scalar implicatures. Linguistics and Philosophy 35. 285–312. DOI:  http://doi.org/10.1007/s10988-012-9119-8

Faller, Martina. 2019. The discourse commitments of illocutionary reportatives. Semantics and Pragmatics 12(8). 1–46. DOI:  http://doi.org/10.3765/sp.12.8

Faller, Martina. 2023. Conjectural speech acts in Cuzco Quechua. Journal of Pragmatics 214. 144–163. DOI:  http://doi.org/10.1016/j.pragma.2023.05.007

Farkas, Donka & Bruce, Kim. 2010. On reacting to assertions and polar questions. Journal of Semantics 27. 81–118. DOI:  http://doi.org/10.1093/jos/ffp010

Ginzburg, Jonathan. 1996. Dynamics and the semantics of dialogue. In Seligman, Jerry & Westerståhl, Dag (eds.), Language, Logic and Computation, vol. 1, 221–237. Stanford: Center for the Study of Language and Information.

Grajales, Róbinson. 2017. La estrategia evidencial dizque en el español de Medellín, Colombia. Onomázein 37. 244–278. DOI:  http://doi.org/10.7764//onomazein.37.02

Greenberg, Yael. 2017. A revised, gradability-based semantics for even. Natural Language Semantics 26. 51–83. DOI:  http://doi.org/10.1007/s11050-017-9140-0

Greenberg, Yael. 2022. On the scalar antonymy of only and even. Natural Language Semantics 30. 415–452. DOI:  http://doi.org/10.1007/s11050-022-09200-x

Groenendijk, Jeroen & Stokhof, Martin. 1984. Studies on the Semantics of Questions and the Pragmatics of Answers. University of Amsterdam dissertation

Gutiérrez, Analía & Matthewson, Lisa. 2012. Evidential determiners: Best (sensory) evidence. Proceedings of Semantics of Underrepresented Languages of the Americas 6. 63–79.

Halliday, M. A. K. 1967. Notes on transitivity and theme in English: Part 2. Journal of Linguistics 3(2). 199–244. DOI:  http://doi.org/10.1017/S0022226700016613

Hamblin, Charles. 1973. Questions in Montague English. Foundations of Language 10. 41–53.

Hanks, William. 2009. Fieldwork on deixis. Journal of Pragmatics 41. 10–24. DOI:  http://doi.org/10.1016/j.pragma.2008.09.003

Hirsch, Aron. 2017. An inflexible semantics for cross-categorial operators. Massachusetts Institute of Technology dissertation.

Huang, James. 1982. Move wh in a language without wh-movement. The Linguistic Review 1(4). 369–416. DOI:  http://doi.org/10.1515/tlir.1982.1.4.369

Izvorski, Roumyana. 1997. The present perfect as an epistemic modal. Proceedings of Semantics and Linguistic Theory 7. 222–239. DOI:  http://doi.org/10.3765/salt.v7i0.2795

Kany, Charles. 1944. Impersonal dizque and its variants in American Spanish. Hispanic Review 12. 168–177. DOI:  http://doi.org/10.2307/469712

Kartunnen, Lauri. 1977. Syntax and semantics of questions. Linguistics and Philosophy 1. 3–44. DOI:  http://doi.org/10.1007/BF00351935

Kayne, Richard S. 1998. Overt vs. covert movement. Syntax 1. 128–191. DOI:  http://doi.org/10.1111/1467-9612.00006

Koev, Todor. 2018. Notions of at-issueness. Language and Linguistic Compass 12. 1–16. DOI:  http://doi.org/10.1111/lnc3.12306

Korotkova, Natasha. 2020. Evidential meaning and (not-)at-issueness. Semantics and Pragmatics 13(4). 1–23. DOI:  http://doi.org/10.3765/sp.13.4

Kovacci, Ofelia. 1999. El adverbio. In Bosque, Ignacio & Demonte, Violeta (eds.), Gramática descriptiva de la lengua española, vol. I, 705–786. Madrid: Real Academia Española, Espasa-Calpe.

Krifka, Manfred. 1992. A compositional semantics for multiple focus constructions. In Jacobs, Joachim (ed.), Informationsstruktur und Grammatik, 17–53. Opladen: Westdeutscher Verlag. DOI:  http://doi.org/10.1007/978-3-663-12176-3_2

López Izquierdo, Marta. 2006. L’émergence de dizque comme stratégie médiative en espagnol médiéval. Cahiers d’études hispaniques médiévales 29. 483–493. DOI:  http://doi.org/10.3406/cehm.2006.1980

Magaña, Elizabeth, 2005. El paso de “dice que” a “dizque”, de la referencia a la evidencialidad. Contribuciones desde Coatepec 8. 59–70.

Martínez Vera, Gabriel. 2020a. On the Semantics of Evidentials in Southern Aymara. University of Connecticut dissertation.

Martínez Vera, Gabriel. 2020b. On Phase-over-Phase Configurations and phase collapsing: wh-extraction in V + de + CP clauses in Spanish. The Linguistic Review 37(3). 393–431. DOI:  http://doi.org/10.1515/tlr-2020-2049

Martínez Vera, Gabriel. 2023. On evidentiality in American Spanish and Brazilian Portuguese. Borealis – An International Journal of Hispanic Linguistics 12. 37–73. DOI:  http://doi.org/10.7557/1.12.1.6716

Martínez Vera, Gabriel. 2024. Direct evidentiality and discourse in Southern Aymara. Natural Language Semantics 32. 1–34. DOI:  http://doi.org/10.1007/s11050-023-09220-1

Matthewson, Lisa & Rullman, Hotze & Davis, Henry. 2007. Evidentials as epistemic modals: Evidence from St’át’imcets. In van Craenenbroeck, Jeroen (ed.), Linguistic variation yearbook 2007, vol. 7, 201–254. Amsterdam: John Benjamins. DOI:  http://doi.org/10.1075/livy.7.07mat

McCready, Elin. 2015. Reliability in Pragmatics. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/acprof:oso/9780198702832.001.0001

Muñetón Ayala, Mercedes & Dorta Luis, Josefa. 2015. La entonación declarativa e interrogativa en el español colombiano de Medellín: voz femenina vs. masculina. Boletín de Filología 50(2). 103–122. DOI:  http://doi.org/10.4067/S0718-93032015000200005

Murray, Sarah. 2010. Evidentiality and the structure of speech acts. Rutgers University dissertation.

Murray, Sarah. 2017. The semantics of evidentials. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/oso/9780199681570.001.0001

Nakanishi, Kimiko. 2012. The scope of even and quantifier raising. Natural Language Semantics 20. 115–136. DOI:  http://doi.org/10.1007/s11050-011-9077-7

Oda, Hiromune. 2022. The NP/DP-language distinction as a scale and parameters in minimalism. University of Connecticut dissertation.

Olbertz, Hella. 2005. ‘Dizque’ en el español andino ecuatoriano: Conservador e innovador. In Olbertz, Hella & Muysken, Pieter (eds.), Encuentros y conflictos: bilingüismo y contacto de lenguas, 77–94. Madrid/Frankfurt: Iberamericana-Vervuert, Madrid Frankfurt. DOI:  http://doi.org/10.31819/9783865278968-005

Olbertz, Hella. 2007. Dizque in Mexican Spanish: The subjectification of reportative meaning. Rivista di Linguistica 19. 151–172.

Pancheva, Roumyana & Rudin, Deniz. 2019. Speaker discourse roles and the discourse profile of reportative evidentials. Proceedings of the 22nd Amsterdam Colloquium, 327–336.

Plungian, Vladimir A. 2001. The place of evidentiality within the universal grammatical space. Studies in Language 12(1). 349–357. DOI:  http://doi.org/10.1016/S0378-2166(00)00006-0

Quek, Yihui & Hirsch, Aron. 2017. Separating focus meanings and focus forms in Standard and Colloquial Singapore English. Proceedings of the North East Linguistic Society 47. 285–298.

Real Academia Española. Banco de datos (CREA) [en línea]. Corpus de referencia del español actual. <http://www.rae.es> Consulted on 1/31/2024.

Reglero, Lara & Ticio, Emma. 2013. A unified analysis of wh-in-situ in Spanish. The Linguistic Review 30(4). 501–546. DOI:  http://doi.org/10.1515/tlr-2013-0018

Reisinger, D. K. E. & Huijsmans, Marianne & Matthewson, Lisa. 2021. Evidentials in the nominal domain: a Speasian analysis of ʔayʔaǰuθəm determiners. Proceedings of Sinn und Bedeutung 25. 751–768.

Riqueros, José. 2013. Spanish Nominal(ization) Patterns. University of Connecticut dissertation.

Rizzi, Luigi. 1997. The Fine Structure of the left periphery. In Haegeman, Liliane (ed.), Elements of Grammar: Handbook in Generative Syntax, 281–337. Dordrecht: Springer. DOI:  http://doi.org/10.1007/978-94-011-5420-8_7

Roberts, Craige. 1996. Information structure in discourse: towards an integrated theory of pragmatics. OSU Working Papers in Linguistics 49. 91–136.

Rochemont, Michael. 2018. Only syntax. In Hartmann, Jutta & Jäger, Marion & Kehl, Andreas & Konietzko, Andreas & Winkler, Susanne (eds.), Freezing: Theoretical Approaches and Empirical Domains, 264–283. Boston: De Gruyter. DOI:  http://doi.org/10.1515/9781501504266-008

Rooth, Mats. 1985. Association with Focus. University of Massachusetts Amherst dissertation.

Rooth, Mats. 1992. Association with focus or association with presupposition? Ms., Cornell University.

Rooth, Mats. 2016. Alternative Semantics. In The Oxford handbook of information structure, 19–40. New York: Oxford University Press. DOI:  http://doi.org/10.1093/oxfordhb/9780199642670.013.19

Rooryck, Johan. 2019. ‘Recycling’ evidentiality: A research program. In Bağrıaçık, Metin & Breitbarth, Anne & De Clercq, Karen (eds.), Mapping linguistic data essays in honour of Liliane Haegeman. University of Ghent. http://hdl.handle.net/1854/LU-8625919.

Ross, John R. 1967. Constraints on variables in syntax. Massachusetts Institute of Technology dissertation.

Rullmann, Hotze & Matthewson, Lisa. 2018. Towards a theory of modal-temporal interaction. Language 94. 281–331. DOI:  http://doi.org/10.1353/lan.2018.0018

Saito, Hiroaki. 2019. Grammaticalization and the Root & Category Theory. Chicago Linguistics Society 54. 461–476.

Saito, Hiroaki. 2021. Grammaticalization as Decategorization. Journal of Historical Syntax 5(10). 1–24.

Sánchez, Liliana. 1995. Aspectual adjectives and the structure of DP and VP. Probus 7(2). 167–180. DOI:  http://doi.org/10.1515/prbs.1995.7.2.167

Sánchez, Liliana. 2010. The morphology and syntax of topic and focus: Minimalist inquiries in the Quechua periphery. Amsterdam: John Benjamins. DOI:  http://doi.org/10.1075/la.169

Sánchez López, Cristina. 1999. Los cuantificadores: clases de cuantificadores y estructuras cuantificativas. In Bosque, Ignacio & Demonte, Violeta (eds.), Gramática descriptiva de la lengua española, vol. I, 1025–1128. Madrid: Real Academia Española, Espasa-Calpe.

Sanromán Vilas, Bengoña. 2020. Do evidential markers always convey epistemic values? A look into three Ibero-Romance reportatives. Lingua 238. 1–26. DOI:  http://doi.org/10.1016/j.lingua.2020.102832

Simons, Mandy & Tonhauser, Judith & Beaver, David & Roberts, Craige. 2010. What projects and why. Semantics and Linguistic Theory 20. 309–327. DOI:  http://doi.org/10.3765/salt.v20i0.2584

Speas, Peggy & Tenny, Carol. 2003. Configurational properties of point of view roles. In Di Sciullo, Anna Maria (ed.), Asymmetry in grammar. Volume 1: Syntax and Semantics [Linguistik Aktuell/Linguistics Today 57], 315–344. Amsterdam: John Benjamins. DOI:  http://doi.org/10.1075/la.57.15spe

Stalnaker, Robert. 1978. Assertion. In Cole, Peter (ed.), Pragmatics, 315–332. New York: Academic Press. DOI:  http://doi.org/10.1163/9789004368873_013

Stalnaker, Robert. 2002. Common ground. Linguistics and Philosophy 25. 701–721. DOI:  http://doi.org/10.1023/A:1020867916902

Suñer, Margarita. 1994. V-movement and the licensing of argumental wh-phrases in Spanish. Natural Language and Linguistic Theory 12. 335–372. DOI:  http://doi.org/10.1007/BF00993148

Travis, Catherine. 2006. Dizque: A Colombian evidentiality strategy. Linguistics 44(6). 1269–1297. DOI:  http://doi.org/10.1515/LING.2006.041

Uriagereka, Juan. 1988. On government. University of Connecticut dissertation.

Van der Auwera, Johan & Plungian, Vladimir A. 1998. Modality’s semantic map. Linguistic Typology 2. 79–124. DOI:  http://doi.org/10.1515/lity.1998.2.1.79

Vanderveken, Daniel. 1990. Meaning and speech Acts, Vol. 1. Principles of language use. Cambridge: Cambridge University Press.

Villalta, Elisabeth. 2008. Mood and gradability. Linguistics and Philosophy 31(4). 467–522. DOI:  http://doi.org/10.1007/s10988-008-9046-x

Wu, Danfeng. 2022. Syntax of either in either … or … sentences. Natural Language and Linguistic Theory 40. 933–977. DOI:  http://doi.org/10.1007/s11049-021-09523-0

Yalcin, Seth. 2007. Epistemic modals. Mind 116. 983–1026. DOI:  http://doi.org/10.1093/mind/fzm983

Zagona, Karen. 2002. The syntax of Spanish. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511613234