TO SEE AS A MARKER OF CONTEXTUAL SALIENCE IN ROMANCE LANGUAGES: EVIDENCE FROM ACADIAN FRENCH AND BRAZILIAN PORTUGUESE

ABSTRACT

This paper discusses constructions found in Brazilian Portuguese (BP) and Acadian French (AF), in which the equivalents of to see accompanies a second verb in imperative utterances. In these constructions in BP and AF to see emphasizes the command expressed by the other verb. The BP construction can also have an additional interpretation, in which vê ‘see_{IMPERATIVE,2,SINGULAR}’¹has the meaning ‘to verify’. It is proposed that BP constructions can be associated with two different structures. The constructions with the ‘to verify’ meaning are treated as biclausal structures in which the verb ver ‘to see’ selects for a CP headed by the complementizer se ‘if’. As for the analysis of the emphatic order meaning associated to the BP and AF constructions, we adopt the proposals put forth in Speas & Tenny (2003) and Hill (2007, 20 , 2014 ) according to which conversational pragmatics is encoded in syntax as a predicative structure (Speech Act Projection - SAP) above CP. Following these ideas, we analyze BP and AF emphatic order constructions as monoclausal structures, where vê in BP and voir in AF are injunctive pragmatic markers that are externally merged into the SA head in order to encode a pragmatic relation.

pragmatic marker; perception verbs; romance languages; grammaticalization; semantic bleaching

RESUMO

Este artigo discute construções do português brasileiro (PB) e da fala informal do francês acadiano (FA) nas quais formas equivalentes à do verbo ‘ver’ aparecem em enunciados imperativos contendo um segundo verbo. Nessas construções, ‘ver’ enfatiza a injunção expressa por esse segundo verbo. A construção do PB apresenta também uma leitura adicional, na qual a forma vê _IMPERATIVO tem o significado de ‘verificar’. O artigo propõe que as construções do PB podem ser associadas a duas estruturas distintas. As construções com o significado de ‘verificar’ são tratadas como estruturas bioracionais nas quais o verbo ‘ver’ seleciona um CP nucleado pelo complementador se . Com relação à análise do significado de ordem enfática associado às construções do PB e do FA, adota-se as propostas de Speas e Tenny (2003) e de Hill (2007, 2014), segundo as quais a pragmática conversacional é codificada na sintaxe com uma estrutura predicativa (uma projeção associada ao Speech Act ‘Ato de Fala’- SAP) acima de CP. Com base nessas propostas, analisam-se as construções de ordem enfática do PB e do FA como estruturas mono-oracionais, em que vê e voir são marcadores pragmáticos injuntivos inseridos diretamente no núcleo SA para codificar uma relação pragmática.

marcador pragmático; verbos de percepção; línguas românicas; gramaticalização; dessemantização

Introduction

It is well known that perception verbs such as to see can select a wide array of tensed and infinitival complements that are associated to different types of semantic categories (events, propositions, etc.) and that these verbs have different readings (direct, indirect, imaginative), depending on the type of complement they occur with. For instance, whereas the presence of a finite clause induces what is known as an “indirect” interpretation of the perception verb (1), the presence of a non-finite complement induces a “direct” reading (2)²( AKMAJIAN, 1977 ; GUASTI, 1993 ; LABELLE, 1996 ; BOIVIN, 1998 ; FELSER, 1999 ; MILLER; LOWREY, 2003 , among others). These verbs also allow an imaginative reading (3), that is, a perception that takes place in the perceiver’s imagination ( OLSSON, 1976 ; BOIVIN, 1998 ; WILLEMS; DEFRANCQ, 2000 ; RODRIGUES, 2006 ).

However, in some Romance languages, the counterparts of to see have other meanings and uses that are not as extensively studied. For instance, in Acadian French (AF) informal speech and in Brazilian Portuguese (BP), equivalents of to see can appear in injunctive utterances in which they do not convey direct, indirect nor imaginative reading. In these utterances, they seem to emphasize (in some readings) the command expressed by the main verb, as in (4) and (5).

In AF, as in other French varieties, infinitival voir (lit.: ‘to see’) can co-occur with imperative verbs, as in (4); however, in these types of utterances, voir does not have the morphosyntactic and semantic properties of the perception verb from which it seems to be derived (see section “Acadian French voir ”). It will be argued that this use of voir has an emphatic function: it reinforces the imperative value by stating to the hearer that the linguistic or extralinguistic context is such that it requires them to undertake the action denoted by the imperative.

The BP construction in (5) can have the same reading as that in (4) for AF: it can be understood as an emphatic order that takes into account salient information from the linguistic or extralinguistic context. However, this construction can have another interpretation, in which ver ‘to see’ has the meaning ‘to verify’. Thus, (5) is ambiguous and could be uttered in two different contexts, for instance:

(i) in a context where the speaker says “you never listen to me, you never pay attention to my words, but in the end I’m always right” (an emphatic order);

(ii) in a context where the speaker says “the phone call quality is bad; let’s do a test to see if you can hear me”. Hence both readings convey an order, but the former transmits an emphatic order, that requires the hearer to undertake the action denoted by the second verb in the sentence, whereas the latter transmits an order or a request that demands a verification act; the verification meaning is conveyed by the verb ver ‘to see’.

The main goal of this paper is to examine the properties of AF and BP constructions under the emphatic order reading, which can be accounted for by conversational pragmatics. To do so, it will first be necessary to differentiate the two meanings of the BP construction given in (5). We will propose that the BP construction in (5) can be associated with two different structures. The construction with the ‘to verify’ meaning is a biclausal structure in which the verb ver ‘to see’ selects for a CP headed by the complementizer se ‘if’, as in (6):

As for the analysis of the emphatic order meaning associated to (5) and also to the AF construction in (4), we adopt the proposals put forth in Speas and Tenny (2003) and Hill (2007 , 2014 ) according to which conversational pragmatics is encoded in syntax as a predicative structure (Speech Act Projection - SAP) above CP. Following these ideas, we analyze AF and BP emphatic order constructions as monoclausal structures, where voir and vê are injunctive pragmatic markers that encode a direct address and these markers are externally merged into the SA head in order to encode a pragmatic relation.

The paper is organized as follows. The next section presents some background notions that are used in this paper. The third section discusses AF data, examining the morphosyntactic and semantic properties of voir as well as its distribution. The fourth section examines BP data and discusses the differences between the two readings for the construction in (5)⁴. Then, in the fifth section, we offer a comparison between AF and BF constructions. The sixth section presents the theoretical framework in which the analysis is couched and in the seventh section we put forward our proposal; the eighth section presents our final remarks.

Background notions

In this section, we briefly present some key concepts we use to characterize AF and BP utterances examined in this paper. These utterances can be subsumed under the group of directive speech acts⁵, whose illocutionary purpose is “an attempt to get the hearer to do something” ( SEARLE, 1975 , p. 158) or to act in a certain way. Directives involve mainly orders and commands, but more moderate directives can be requests, instructions or suggestions.

Generally, it is the imperative that is used in directive speech acts, but other types of sentences can also fulfill that function. It is also to be noted that imperative sentences can be non-directive ( KISSINE, 2013 ; JARY; KISSINE, 2016 ). Accordingly, even if BP sentences that express an emphatic order are in the indicative mood, we will analyze them as directive speech acts.

In discussing the meaning of imperatives, Aikhenvald (2010 , p. 203) argues that directive meanings “correlate with a general feature of the strength or degree of a command” and that “emphasis in imperatives is often linked to the strength of command – the more insistent the commander, the more emphasis they put on the command”. The author explains that an imperative by itself can convey a neutral order. However, it can be strengthened, i.e. the peremptory nature of the command can be intensified; in that case, “imperatives are expected to be more formally marked than neutral imperatives” ( AIKHENVALD, 2010 , p. 204). To illustrate this property, Aikhenvald points to an example in Haro, an Omotic language from Ethiopia, in which an utterance marked by an emphatic -tte following the imperative expresses a stronger command than the neutral imperative.

Therefore, although the notion of emphasis is an intuitive one, we assume that an emphatic order is a recourse the speaker employs to show the hearer that the context is such that they (the hearer) must comply with the given order. Some available grammatical marking of emphasis in languages would be, for instance, a characteristic prosody, the presence/absence of the subject or the use of markers⁶. We assume that the AF and BP pragmaticalized forms discussed in this paper are part of the range of resources that can mark the emphasis in a directive speech act. Since our main objective is to discuss this emphatic reading in contrast with a more neutral use of imperatives, we will not distinguish, when talking about this more neutral use, orders or commands from requests or suggestions, using these terms indistinctively.

As for the concept of pragmaticalization, following Dostie (2004 , 2009 ), we take it to be a process similar to grammaticalization, but with an outcome that bears a pragmatic function instead of a more grammatical one. It is to be noted that the term grammaticalization is often used in the literature to refer to pragmaticalization. Grammaticalization is a process of linguistic change by which lexical items become grammatical items, and grammatical items achieve an even more grammatical status ( HOPPER; TRAUGOTT, 2003 ; ROBERTS; ROUSSOU, 2003 ; HEINE; NARROG, 2012 , among others). Heine and Narrog (2012) propose a set of four parameters to be used in the identification and description of grammaticalization instances: extension, desemanticization, decategorialization and erosion. The extension parameter refers to the emergence of new meanings from the extension of linguistic expressions to new contexts; the desemanticization parameter, also known as semantic bleaching , refers to loss in semantic content; the parameter of decategorialization concerns loss of morphosyntactic properties; and erosion refers to loss of phonetic substance.

We will argue that AF and BP utterances conveying an emphatic order involve a pragmatic marker – respectively voir and vê – that underwent a process of pragmaticalization starting from the lexical verb forms ( voir and ver ), by having been extended to new contexts and having lost semantic content and morphosyntactic properties. Consequently, our view is that this process is diachronic; however, an explanation for the path of semantic change from lexical meaning to pragmatic meaning for those markers is beyond the objectives of the current study. Furthermore, whether or not there remains a discernible semantic relation between the lexical verb and the pragmatic marker does not interfere with our description of the data⁷.

Acadian French voir

In this section, we provide a sketch of voir in Acadian French as it appears in spontaneous (informal) imperative utterances. As shown in (7), voir most often pronounced [wɛ:r] in this dialect can be used in both affirmative and negative imperative constructions.

These constructions are attested in different areas of the Atlantic provinces (Canada): in Nova Scotia ( BOUDREAU, 1988 ; HENNEMAN; NEUMANN-HOLZCHUH, 2014), in New Brunswick ( POIRIER, 1993 ; WIESMATH, 2006 ) and Newfoundland and Labrador ( BRASSEUR, 2001 ). They are also in usage in Louisiana ( DAIGLE, 1984 ; PAPEN; ROTTET, 1997 ; PARR, 1940 ; VALDMAN et al., 2010).

Regarding their morphological properties, in these types of examples, voir does not have the typical behavior of verbs in general. Indeed, in imperative utterances, voir is invariable; it cannot be inflected (8a), which contrasts with the visual perception verb, that can bear all types of inflections for tense, mood and person (8b).

The distributional restrictions of Acadian French voir shows that this item does not have syntactic flexibility, much like clitics: it almost always has to follow the imperative verb.

Nominal phrases, either complements (9) or adjuncts (10), cannot be inserted between the conjugated verb in the imperative and voir .

Also, as with clitics, the negation element pas cannot intervene between voir and the verb, as shown in (11).

It seems that only certain pronouns, that is, clitics, can be inserted between the imperative and voir .

The data in (12) can be accounted for easily by the fact that one hallmark property of clitics is that they lack autonomy or at least their autonomy is reduced; hence, they have to appear next to the verb. As can be seen from the data discussed above, the scope of voir is very narrow: it is limited to the imperative verb.

In regard to the semantic properties of these structures, as shown in (13), in imperative constructions, voir can co-occur with a perception verb (for instance, a visual or auditory perception verb), which shows that the use of this form is accompanied by the loss of characteristics associated to the perception verb. In other words, in these contexts, voir is characterized by desemantization.

Moreover, voir in these utterances can be omitted (14), which suggests that it does not add semantic content at the sentence level (or at least very little); rather, it has a function at the discourse level; in other words, it has a pragmatic function.

Example in (15) further shows that voir in imperative contexts does not behave as a full-fledged verb; it cannot have arguments in contrast to the visual perception verb that can select different types of complements (for instance, nominal phrases and indicative subordinate clauses).

In (15a) and (15b), voir does not select the nominal phrase nor the indicative complement, which are arguments of the conjugated verb imagine . The fact that voir is not obligatory in these utterances provides support for this claim. However, voir in its use as a perception verb can take different types of complements, as illustrated in (15c) and (15d). If we hypothesize that voir in imperative contexts is derived from the perception verb, then we must conclude that it has lost its argument structure.

As shown in the imperative utterances discussed above, voir does not have the morphosyntactic and semantic/pragmatic properties of the perception verb from which it seems to be derived. Concerning the pragmatics of these constructions, it is argued that this use of voir has an emphatic function: it reinforces the imperative value by stating to the hearer that the context (linguistic or extralinguistic) is such that it requires them to undertake the action denoted by the imperative. For instance, for (16), voir can be linked to a linguistic antecedent or an extralinguistic antecedent.

An example of a case where voir can be associated to a linguistic antecedent is linked in (17). Let us imagine a situation in which a couple is arriving at their home, after getting groceries. In this example, a request was previously formulated in the discourse by the speaker, but was ignored by the hearer. The use of the imperative with voir allows the speaker to signal to the hearer that they should base themselves on the context, which should be mutually salient because the utterance was just pronounced, and also to insist on their demand.

This example would also be perfectly acceptable if no linguistic context was available, as long as the speaker has access to sufficiently tangible or obvious information in the extralinguistic context. In these types of cases, the antecedent would not be formally uttered; it would be implicit from the extralinguistic context. Hence, Ferme voir la porte! ‘Close the door!’ could be uttered in cases where it is evident that there is too much noise, that the speaker has a secret to confide in and wants to discuss in private, that the speaker is cold, that their hands are full and that they are not able to close the door themselves. Crucially, the utterances with voir , as the one in (18), are only appropriate if the antecedent, be it a linguistic or an extralinguistic antecedent, is mutually obvious. Consequently, it would be considered bizarre to utter (18) in the case where no linguistic antecedent is available (if no previous request has been done) or in the case where, for instance, the utterance is pronounced by a parent and is addressed to a child who is disciplined, studious, and punctual in their work. In these cases, the utterance would be incoherent or infelicitous.

In a nutshell, while an utterance such as (19a), where voir is absent, is used simply to make a request, (19b), with voir , highlights the fact that the hearer should infer from the linguistic or extralinguistic context that they should close the door. Therefore, (19b), but not (19a), renders salient the fact that the circumstances require the hearer to perform the specific action expressed by the imperative verb.

In these contexts, voir is similar to donc (lit.: so ) in some of its uses in other varieties of French (see e.g., DOSTIE, 2004 ; VINCENT, 1993 ; VLEMINGS, 2003 ).

Brazilian Portuguese vê

In this section, we will discuss the ambiguous data from BP exemplified in (5), repeated here as (20) for convenience.

As mentioned in the introduction, this construction can have the same emphatic order reading as the AF constructions discussed in the previous section. However, it can also have a reading in which the perception verb means ‘to verify’. In order to compare AF and BP, we must first differentiate the two readings that BP constructions such as the one in (20) can receive. In this section, we address specifics of these readings and discuss their main properties.

The verification order reading in (20) could be uttered, for instance, in a context where the speaker says “the phone call quality is bad; let’s do a test to see if you can hear me”. On the other hand, with the emphatic order reading, (20) could be uttered in a context where the speaker says “you never listen to me, you never pay attention to my words, but in the end I’m always right”.

These two readings can thus be described as injunction readings; however, the sentences do not convey the same orders. In (20), for example, under the ‘to verify’ meaning, the speaker’s intention is to get the addressee to verify if they (the speaker) can be heard; it is not to get the addressee to pay attention to what they are saying, that is, to the order given in relation to the linguistic or extralinguistic context. For the ‘to verify’ interpretation, the order to be executed is transmitted by the perception verb (which has in this case the meaning ‘to verify’ or ‘to determine’). (21a) would be a paraphrase of this reading. A possible reply for (20) could be (21b), with the perception verb, but not (21c), with the verb escutar ‘to listen’.

Under the emphatic order meaning, on the contrary, although the form ver ‘to see’ in (20) corresponds to the 2^ndperson singular of the true imperative, it does not convey an order to get the addressee to verify or to determine something; it communicates an emphatic order for the execution of the action denoted by the second verb. For instance, (22) is a paraphrase of that meaning in (20). A possible reply for (20) would be (22b), with the second verb, not (22c), with the perception verb.

Despite the fact that (22a) is a paraphrase of (20) in the case of the emphatic order meaning, there is a pragmatic difference between these constructions. While the sentence in (22a) is used to make a request, to give an order or simply to draw the hearer’s attention to something the speaker thinks is important, the sentence in (20), under the emphatic order reading, emphasizes that the hearer should take the context into account and execute the specific action denoted by the second verb. It may be the case that a specific request has already been made before but was not heard (or ignored completely by the addressee) or it may be that there is salient information available from the context to both the speaker and the hearer. For instance, if A says to B “you never listen to me, you never pay attention to my words, but in the end I am always right”, and then utters (20), what A is implicitly stating is that B should really listen to them this time, because if B does not do so, as it happened in previous situations, B will probably encounter the same problems as before. This context could also be implicit and mutually obvious in the conversational situation.

In the next sections, we will see that differences between the emphatic order meaning and the ‘to verify’ meaning are correlated with morphosyntactic distinctions.

In the verifying reading, the perception verb displays a full paradigm, as illustrated in (23), with finite forms, and in (24), with non-finite forms. In fact, as shown by the examples, the imperative (23a), the present (23b) and the simple past (23c) forms are accepted. The perception verb can also be in the infinitive form (24a), appearing in the periphrastic future structure composed of the verb go in the present indicative form and the verb see in the infinitive; in the gerund form (24b), appearing in the past continuous structure composed of the verb be in the past imperfect and the verb see in the gerund form; and in the participle form (24c), appearing in the past perfect structure composed of the verb have in the past imperfect form and the verb see in the participle form.

In the emphatic order reading, on the contrary, the perception verb has a fixed form, which is vê . As shown in (25) and (26), it cannot bear any other finite forms and it cannot bear non-finite forms. In this interpretation, ver displays instead a defective paradigm, permitting only one form, which is vê , imperative, second person, singular, as in (27).¹⁰

When ver receives the verifying reading, it is a bivalent verb: two arguments are required with this predicate, an agent and a theme, as shown in (28): the theme is realized by a clause in (28a, b), and by a nominal phrase in (28c); the agent is realized by pro in (28a) and by the pronoun ela ‘she’ in (28b). Examples in (28d, e) show that these two arguments are required. It is also possible to insert a benefactive PP (31).

On the other hand, in the emphatic order reading, vê is a fixed form [imperative, 2^ndperson singular] which can only combine with a [ se ...] constituent (30). Example in (31) shows that it is impossible to insert a benefactive PP.

Acadian French vs. Brazilian Portuguese

The two previous sections described constructions from AF and BP used to convey emphatic orders.¹¹These constructions involve a form of the verb equivalent of to see: voir in AF and vê in BP. As described in detail in sections “Acadian French voir ” and “Brazilian Portuguese vê ”, these forms do not have the morphosyntactic and semantic properties of the perception verb from which they seem to be derived. In contrast with the lexical verbs voir (AF) and ver (BP) ‘to see’, the forms voir and vê do not behave as full-fledged verbs in that they are invariable, fixed forms: in AF it only appears in the infinitive and in BP the 2^ndperson singular imperative form is the only one possible.¹²Moreover, these forms in both AF and BP lack an argument structure.

Their syntactic properties, however, are not the same. In AF constructions, we have an imperative sentence, and voir seems to exhibit characteristics akin to clitics, in the sense that it has to follow the imperative verb, unless it co-occurs with clitic pronouns. In BP, the sentence is in the indicative mood; the form vê does not behave like a clitic pronoun, unlike in AF, since it is not bound to the lexical verb; rather, this form seems to select a [ se… ] complement.

Despite these syntactic differences, AF and BP forms have the same pragmatic function: they emphasize an order, signaling to the addressee that they must take the (linguistic or extralinguistic) context into account and execute the action denoted by the lexical verb. In other words, these constructions can only be uttered if some contextual information is obvious for both the speaker and the hearer¹³.

As we will discuss in more detail in “Analysis”, we analyze the AF form voir and the BP form vê as pragmatic markers and we propose a similar syntactic analysis for both constructions, which is couched in the theoretical framework of Speas and Tenny (2003) and Hill (2007 , 2014 ). The next section provides information on these studies.

Theoretical framework

Several studies on the cartographic project ( RIZZI, 1997 , 2013 ; CINQUE 1999 ; CINQUE; RIZZI, 2012 ; RIZZI; CINQUE, 2016 ) argue that the pragmatic roles of speaker and hearer are syntactically encoded at the left periphery of clauses ( SPEAS; TENNY, 2003 ; HILL, 2007 , 2014 ; HAEGEMAN, 2014 ; MIYAGAWA, 2012 ). Their proposals intend to explain phenomena like vocatives, allocutive agreement, logophoricity, speaker-oriented particles, and grammaticalized adverbs, among others.

According to Speas and Tenny (2003) and Hill (2007 , 2014 ), conversational pragmatics is encoded in syntax in a Speech Act Phrase (SAP), a performative predicative structure above the CP domain. This proposal is a revival of Ross’ (1970) Performative Hypothesis, according to which even declarative clauses would be performative. According to Ross, in Deep Structure (the input of interpretation), a declarative clause would be embedded in another clause containing a performative verb encoding the illocutionary declarative force of the utterance. The surface structure would be derived after the performative deletion rule was applied to delete the embedding clause.

Speas and Tenny’s SAP structure, on the contrary, does not involve a performative verb; it consists of two layers that extends the functional projection of the clause, as illustrated in (32). Speas and Tenny, following Hale and Keyser’s (1999) ideas, argue that SA structure obeys the same restrictions found in argument structure. Hence, replicating the v P structure for the saturation of theta-roles, the SA head projects a structure that would ensure the saturation of pragmatic roles, with two head positions, two specifiers and one complement.

The authors argue that what is represented are not the speech acts as such, but the asymmetric relations among the pragmatic roles speaker, hearer, utterance content . This structural configuration would explain the restrictions observed on the number of grammaticized speech acts across languages.

On the basis of Romanian data, Hill (2014 , p.147) proposes the modified representation given in (32) illustrated in (33).

Hill argues that the Romanian speech act particles provide lexical evidence for the sa head. These pragmatic particles entail a predicative interpretation, which leads to the prediction that this syntactic behavior follows from their merging at some level of SAP.

However, according to Hill, Romanian data suggest that these particles do not select hearers , but CPs ( utterance content ) headed by lexical complementizers. Hence, it would not be possible to propose saP to replace ForceP , since Force , being occupied by a complementizer, could not spell out the speech act head. In Hill’s representation, ForceP , associated to the utterance role, is the complement of sa , while the hearer pragmatic role is saturated in Spec, saP , as are indirect objects in the v P. In Hill’s (2014 , p. 147) own words:

[...] there is a speech act head SAh that behaves as a verb insofar as it merges with a direct complement (ForceP) and projects a phrase structure that allows for the merging of a constituent in the hearer position (Spec, SAhP); the derivation extends with the speaker field (SAsP), where the speaker p-role is checked, possibly by a constituent merged in Spec, SAsP.

Analysis

As we have seen above, the AF construction in (4) and the BP construction in (5) can have the same emphatic order reading. Under this reading, AF voir and BP vê share some important properties, such as morphological invariability, and lack of argument structure; furthermore, they have the same pragmatic function (see section “Acadian French vs. Brazilian Portuguese”). These properties show that these elements underwent a process of decategorialization and desemantization, acquiring a pragmatic function, which is typical of the pragmaticalization process. As Dostie (2004 , 2009 ), we take pragmaticalization to be a process similar to grammaticalization, but with an outcome that bears a pragmatic function instead of a more grammatical one. Hence AF and BP forms do not function as lexical verbs heading a matrix clause, but rather behave as pragmatic markers, emphasizing an order or a request. As we have discussed, the differences between AF constructions in (34) and between BP constructions in (35) can be accounted for in pragmatic terms. While the (a) utterances are simple requests or simple orders, the (b) utterances highlight the fact that the hearer should rely on the context which provides information that warrants that the situation should be executed.

Our hypothesis is that AF and BP emphatic order constructions are monoclausal structures, where voir and vê function as discourse markers externally merged in higher domains: SAP ( HILL, 2007 , 2014 ). Below we discuss the specifics of these structures.

Moreover, we have seen that BP constructions can have another interpretation, which we dubbed the ‘to verify’ meaning. In this case, vê presents some properties that indicate it is a lexical verb and heads the matrix clause of a biclausal structure: it displays a full morphological paradigm and a full argument structure – an external argument (DP); an internal argument ([_ForceP se ... ] or DP) and an optional argument (benefactive PP). We present a detailed analysis for this reading below.

We argue that the speech act markers voir in AF and vê in BP are externally merged into SA_h, the head that encodes a pragmatic relation to the hearer, to indicate what they have to do in view of the (extra)linguistic context (36).

The structure’s derivation for the AF sentence in (37a) is shown in (37b-c).

The structure’s derivation for the BP sentence in (38a) is shown in (38b-c). We argue that the pragmatic marker vê in BP utterances expressing an emphatic order is derived from the lexical verb ver ‘to see’ in its use with a se -clause as complement. This functional element is merged in SA_Hand selects a CP headed by the complementizer se ¹⁴.

This analysis predicts that vê cannot be negated because it occupies a position higher in the syntactic representation than the position for negation. This prediction is borne out, as shown in (39) for BP¹⁵.

Our hypothesis to account for the verifying meaning of BP sentences is that they are biclausal structures; in this use, vê is a lexical verb, with the structure in (40).

This structure predicts that there are two temporal domains in the sentence and that both verbs can be negated. Data in (41) and (42) show respectively that these predictions are borne out.

These data show that the range of meanings generally associated with perception verbs, especially to the verb ‘to see’, must be extended. In addition to the meanings of direct, indirect and imaginative perception, it is possible to include the meaning of verifying perception (perception with a verification, inquiring, checking objective). We argue that these data corroborate Labelle’s (1996) and Boivin’s (1998) hypothesis that the different readings of ‘to see’ are associated to a single lexical entry, which codifies an abstract and underspecified meaning (‘to perceive’). This underspecified meaning would become more precise according to the complement the perception verb is associated with. Therefore, the verb ‘to see’ is not itself ambiguous: its multiple interpretations are the result of its association with distinct types of complements. These different structures would realize the semantic selection of ‘to see’, which Labelle (1996) terms “Object of Perception”. Labelle argues that, given the multiple semantic categories that can be selected by this verb (Proposition, Event, Action, Object), it is preferable to subsume them under a more general one. Hence, every construction having a compatible semantic type with “Object of Perception” could be a complement of ‘to see’.

In the cases discussed in this paper, the verification meaning of ‘to see’ would be derived from the more general meaning ‘to perceive’. Considering that the definition for ‘to verify’ as “to establish the truth, accuracy, or reality of” ( Merriam-Webster , online), it is possible to argue that, to state the truth of a situation, this situation must be previously perceived. Therefore, in a sentence like (43), the content expressed by the se- clause can be true or false, since the complementizer se introduces an interrogative clause, whose truth value is open ( RODRIGUES; LUNGUINHO, 2017 ). It is hence this “open truth value situation” of the se- clause that specifies the meaning of ‘to see’ as ‘verification’, ‘inquiry’.¹⁶

Figure 1 summarizes the analysis for the constructions studied.

Figure 1
– Analysis

Final remarks

In this paper, we described the properties of voir (Acadian French) and vê (Brazilian Portuguese) used to convey an emphatic order. We argued that these forms are pragmaticalized elements, i.e. injunctive pragmatic markers externally merged on SA_HP ( HILL, 2007 , 2014 ), and that the sentences in which these markers appear are the realization of different operations applied to the same abstract structure of (functional) heads.

We also showed that vê in constructions with the form [ vê + se ... ] in Brazilian Portuguese can be associated with: a) an emphatic order meaning or b) a verification meaning. Despite being superficially similar, they are semantically and syntactically different:

1. when vê expresses an emphatic order, [ vê se ...] is a monoclausal structure and vê is a SA_Hhead;

2. when vê express a verify meaning, [ vê se ...] is a biclausal structure and vê is a lexical head.

REFERÊNCIAS

• AIKHENVALD, A. Y. Imperatives and commands. Oxford: Oxford University Press, 2010.
• AKMAJIAN, A. The complement structure of perception verbs in an autonomous syntax framework. In: CULICOVER, P. W.; WASOW, T.; AKMAJIAN, A. (org.). Formal syntax. New York: Academic Press, 1977. p. 427-460.
• BOIVIN, M. C. Complementation and interpretation: The concrete and imaginative readings of ‘visual’ perception verbs. MIT Working Papers in Linguistics, Cambridge, n. 25, p. 103-123, 1998.
• BOUDREAU, É. Glossaire du vieux parler acadien: mots et expressions recueillis à Rivière-Bourgeois (Cap-Breton). Montreal: Éditions du Fleuve, 1988.
• BRASSEUR, P. Dictionnaire des régionalismes du français de Terre-Neuve. Tübingen: Niemeyer, 2001.
• CINQUE, G. Adverbs and functional heads: A cross-linguistic perspective. Oxford: Oxford University Press, 1999.
• CINQUE, G.; RIZZI, L. The cartography of syntactic structures. In: HEINE, B.; NARROG, H. (org.). The Oxford Handbook of Linguistic Analysis. Oxford: Oxford University Press, 2012. p. 51-65.
• CRUSCHINA, S. The expression of evidentiality and epistemicity: Cases of grammaticalization in Italian and Sicilian. Probus, Berlin, v. 27, n. 1, p. 1-31, 2015.
• CRUSCHINA, S.; REMBERGER, E.-M. Speaker-oriented syntax and root clause complementizers. Linguistic Variation, Amsterdam, v. 18, n. 2, p. 336-358, 2018.
• DAIGLE, J. O. A dictionary of the Cajun language. Ann Arbor (Michigan): Edwards Brothers Inc., 1984.
• DOSTIE, G. Discourse markers and regional variation in French: A lexico-semantic approach. In: BEECHING, K.; ARMSTRONG, N.; GADET, F. (org.). Sociolinguistic variation in contemporary French. Amsterdam; Philadelphia: John Benjamins, 2009. p. 201-214.
• DOSTIE, G. Pragmaticalisation et marqueurs discursifs: Analyse sémantique et traitement lexicographique. Brussels: De Boeck: Duculot, 2004.
• FELSER, C. Verbal complement clauses. Amsterdam: John Benjamins, 1999.
• FISCHER, K. (org.). Approaches to discourse particles. Oxford: Elsevier, 2006.
• GUASTI, M. T. Causative and perception verbs: A comparative study. Torino: Rosenberg & Sellier, 1993.
• HAEGEMAN, L. West flemish verb-based discourse markers and the articulation of the speech act layer. Studia Linguistica, Chichester, n. 68, p. 116-139, 2014.
• HALE, K.; KEYSER, J. Bound features, merge and transitivity alternations. In: PYLKKÄNEN, L.; VAN HOUT, A.; HARLEY, H. (org.). Papers from the UPenn/MIT Roundtable on the Lexicon. Cambridge: MIT Press, 1999. p. 49-72. (MIT Working Papers in Linguistics, 35).
• HEINE, B.; NARROG, H. Grammaticalization and linguistics analysis. In: HEINE, B.; NARROG, H. (org.). The Oxford Handbook of Linguistic Analysis. Oxford: Oxford University Press, 2012. p. 401-423.
• HENNEMANN, J.; NEUMANN-HOLZSCHUH, I. Les particules voir et –ti dans le français acadien et louisianais : deux particules à cheval entre lexique et syntaxe. In: ARRIGHI, L.; LEBLANC, M. (org.). La francophonie en Acadie: Dynamiques sociales et langagières: Textes en hommage à Louise Péronnet. Sudbury: Prise de parole, 2014. p. 107-134.
• HILL, V. Vocatives: How syntax meets with pragmatics. Leiden: Brill Publishers, 2014.
• HILL, V. Vocatives and the pragmatics-syntax interface. Lingua, Amsterdam, n. 117, p. 2077-2105, 2007.
• HOPPER, P. J.; TRAUGOTT, E. C. Grammaticalization. Cambridge: Cambridge University Press, 2003.
• JARY, M.; KISSINE, M. When terminology matters: The imperative as a comparative concept. Linguistics, Berlin, v. 54, n. 1, p. 119-148, 2016.
• KISSINE, M. From utterances to speech acts. Cambridge: Cambridge University Press, 2013.
• LABELLE, M. Remarques sur les verbes de perception et la sous-catégorisation. Recherches linguistiques de Vincennes, Saint-Denis, n. 25, p. 83-106, 1996.
• MILLER, P.; LOWREY, B. La complémentation des verbes de perception en français et en anglais. In: MILLER, P.; ZRIBI-HERTZ, A. (org.). Essais sur la grammaire comparée du français et de l’anglais. Paris: Presses Universitaires de Vincennes, 2003. p. 131-188.
• MIYAGAWA, S. Agreements that occur mainly in the main clause. In: AELBRECHT, L.; HAEGEMAN, L.; NYE, R. (org.). Main clause phenomena: New Horizons. Amsterdam: John Benjamins, 2012. p. 79-112.
• OLSSON, K. La construction: verbe + objet direct + complément prédicatif en français: Aspects syntaxiques et sémantiques. 1976. PhD Thèse (Docteur en Linguistique) - Université de Stockholm, Stockholm, 1976.
• PAPEN, R. A.; ROTTET, K. A structural sketch of the Cajun French spoken in Lafourche and Terrebonne parishes. In: VALDMAN, A. (org.). French and Creole in Louisiana. New York: Plenum, 1997. p. 71-108.
• PARR, U. M. A glossary of the variants from standard French in Terrebonne Parish with an appendix of popular beliefs, superstitions, medicines, and cooking recipes. 1940. Dissertation (Master of Arts) - Louisiana State University, Baton Rouge, 1940.
• POIRIER, P. Le glossaire acadien. Critical edition by Pierre M. Gérin. Moncton: Éditions d’Acadie: Centre d’études acadiennes, 1993.
• RIZZI, L. Notes on cartography and further explanation. Probus, Berlin, n. 25, p. 197-226, 2013.
• RIZZI, L. The fine structure of the left periphery. In: HAEGEMAN, L. (org.). Elements of grammar. Dordrecht: Kluwer, 1997. p. 281-337.
• RIZZI, L.; CINQUE, G. Functional categories and syntactic theory. Annual Review of Linguistics, Palo Alto, v. 2, p. 139-163, 2016.
• ROBERTS, I.; ROUSSOU, A. Syntactic change: A minimalist approach to grammaticalization. Cambridge (MA): Cambridge University Press, 2003.
• RODRIGUES, P. Les compléments infinitifs et gérondifs des verbes de perception en portugais brésilien. 2006. PhD Thèse (Docteur en Linguistique) - Université du Québec à Montréal, Montreal, 2006.
• RODRIGUES, P.; LUNGUINHO, M. V. Ver como marcador pragmático em português brasileiro. Revista Linguística, Rio de Janeiro, v. 13, n. 2, p. 231-262, 2017.
• ROSS, J. R. On declarative sentences. In: JACOBS, R. A.; ROSENBAUM, P. S. (org.). Readings in English transformational grammar. Washington, DC: Georgetown University Press, 1970. p. 222-272.
• SCHERRE, M. M. P.; CARDOSO, D. B. B.; LUNGUINHO, M. V.; SALLE, H. Reflexões sobre o imperativo em português. Delta, São Paulo, v. 23, n. esp., p. 193-241, 2007.
• SEARLE, J. R. A taxonomy of illocutionary acts. In: GUNDERSON, K. (org.). Language, mind and knowledge. Minneapolis: University of Minnesota Press, 1975. p. 344-369.
• SPEAS, M.; TENNY, C. Configurational properties of point of view roles. In: DI SCIULLO, A.M. (org.). Asymmetry in grammar. Amsterdam: John Benjamins, 2003. p. 315-344.
• VALDMAN, A.; ROTTET, K. J.; ANCELET, B. J.; GUIDRY, R.; KLINGLER, T. A.; LAFLEUR, A.; LINDNER, T.; PICONE, M. D.; RYON, D. Dictionary of Louisiana French: As Spoken in Cajun, Creole, and American Indian Communities. Jackson: The University Press of Mississippi, 2010.
• VINCENT, D. Les ponctuants de la langue et autres mots du discours. Quebec City: Nuit Blanche, 1993.
• VLEMINGS, J. The discourse use of French donc in imperative sentences. Journal of Pragmatics, Amsterdam, n. 35, p. 1095-1112, 2003.
• WIESMATH, R. Le français acadien: analyse syntaxique d’un corpus oral recueilli au Nouveau-Brunswick/Canada. Paris: L’Harmattan, 2006.
• WILLEMS, D.; DEFRANCQ, B. L’attribut de l’objet et les verbes de perception. Langue française, Malakoff, n. 127, p. 6-20, 2000.

• 1
  The following abbreviations are used hereafter in this paper: 1=1 person; 2=2 person; 3=3 person; expl=expletive; FUT=future; GER=gerund; IMP=imperative; IMPSURROGATE=surrogate imperative; IMP_TRUE=true imperative; INF=infinitive; PART=participle; PRES.IND=present of the indicative mood; PSTIMP=past imperfect; PSTPERF=past perfect; sg=singular; pl=plural; SPPST=simple past; SUBJ=subjunctive mood.
• 2
  The labels direct and indirect are related to the experiencer’s contact with the perceived event. Indirect perception is generally a result of an inferential activity based on evidence related to the described situation (in (1), the way Jean was mumbling his words, for instance). Direct perception, on the contrary, cannot be based on the perception of this evidence (2).
• 3
  The form escuta is in the indicative present, second person singular; it is not the true imperative form of escutar ‘to listen’. The sentence in (i), in which the verb form is plural, confirms this analysis: (i) Vê se vocês me escutam desta vez ! see-2sg-IMP if you me listen-2pl-PRES.IND
• 4
  This discussion is based on Rodrigues and Lunguinho (2017) .
• 5
  For a characterization of speech acts, see, for instance, Searle (1975) .
• 6
  For an extensive discussion about grammatical marking of strength of command, see Ainkhenvald (2010, section 6.3)
• 7
  For an overview about the discussion on the polyfunctionality of discourse markers and the relationship between their meanings as a case of homonymy or polysemy, see Fischer (2006) .
• 8
  The symbol # indicates that the utterance is well formed, but does not have the intended meaning. The utterances that are preceded by # in this article are used to dare someone to do something. In these cases, voir is used to express somewhat a threat and has retained the meaning of the visual perception verb. These utterances seem to be used elliptically, as in Goûte ça (pour) voir (ce qui se passera)! ‘Taste that (in order to) see (what will happen)!’.
• 9
  In BP, there is only one true form of the imperative, the second person singular; this true form can alternate with a surrogate subjunctive form. Unlike in European Portuguese and Spanish, the true imperative in BP, as well as the surrogate one, is compatible with negation (SCHERRE et al. , 2007).
• 10
  The surrogate form for the imperative (i) is accepted for some speakers. (i) %Veja se (você) come. see-2sg-IMPSURROGATE if (you) eat-3sg-PRES.IND
• 11
  We will put aside in this section the other meaning of BP constructions.
• 12
  See note 4.
• 13
  See the discussion about example (16) and the contextual information.
• 14
  On functional elements that select a clause, see Cruschina (2015) , Cruschina and Remberger (2018) and Hill (2014)
• 15
  See note 6.
• 16
  In sentences like (i), (ii) and (iii), whose complements are indirect interrogatives, ‘to see’ also has a verifying meaning. (i) Vê como a Maria pagou a conta da farmácia. see how the Maria paid the drugstore bill ‘See how Maria paid the drugstore bill.’ (ii) Vê por que a Maria ainda não chegou. see why the Maria yet not arrived ‘See why Maria hasn’t arrived yet.’ (iii) Vê quando a Maria vai chegar. see when the Maria will arrive ‘See when Maria will arrive’.

Publication Dates

• Publication in this collection
  13 Dec 2021
• Date of issue
  2021

History

• Received
  21 May 2020
• Accepted
  22 Aug 2020