Supportive moves in requests and orders in brazilian portuguese and uruguayan spanish variant

Considering the linguistic politeness studies (Brown & Levinson, 1987) and the request analysis categories usually described in cross-cultural research (Blum-Kulka, House, & Kasper, 1989a), this paper examines the discursive strategies that support requests and orders in a corpus produced by brazilians (from Curitiba), and uruguayans (from Montevideo). It is sought to verify whether the traditionally described categories apply to the data and to what extent the strategies used agree in these two linguisticcultural communities. To this end, different contextual variables are taken into account by means of a Written Discourse-Completion Test. We raised the initial hypothesis that geographical and cultural nearness between the two societies favors a strong similarity in the strategies used by their members. However, the comparison between the external mitigating strategies produced in specific contexts has shown certain pragmatic and linguistic patterns that are specific of each group, as it is the case of the increased tendency towards the expression of negative politeness by uruguayan informants.


Introduction
From the beginning of the 1980's, empirical studies in the area of intercultural pragmatics have proposed to identify which aspects are universal, and which ones are specific to each culture, in the production of distinct speech acts, taking into account, as well, their relation with linguistic politeness.In this context, considerable attention has been given to strategies employed by speakers in directive acts, such as requests and orders.This is because the nature of such acts offers a fecund ground to the study of potential connections between linguistic politeness, and cultural and social variables, such as the degree of familiarity between speakers and power relationships.Considering as reference the linguistic politeness model of Brown and Levinson (1987) and the categories of analysis described in Blum-Kulka, House and Kasper (1989a), in this work we examine the discursive strategies that support requests and orders in a corpus generated by brazilian subjects from the City of Curitiba, and uruguayan subjects from Montevideo.By means of a qualitative and quantitative analysis, we try to find out whether the traditionally described categories are applicable to the data, and in what extent the strategies used coincide or not with the two linguistic-cultural communities.For this, we take into account variables such as the kind of request/order, the kind of social relationship existing among the participants, the relative distance and the power relationships established among them, from the roles played in each situation.Our hypothesis is that both geographic and cultural proximity favors a strong similarity in the discursive strategies employed in the two communities.
The existing literature on verbal politeness has examined empirically several aspects of requests in many languages and in different variants of spanish.Among the contrastive studies on this issue, one finds: american english and hebrew (Blum-Kulka, 1987); french canadian, australian english, german, hebrew and argentinian spanish (Blum-Kulka et al., 1989a); turkish and german (Marti, 2006); american english and chinese (Lin, 2009); English, german, polish and russian (Ogiermann, 2009); ecuadorian spanish and iberian spanish (Placencia, 1998); cuban spanish (Ruzicková, 2007); mexican spanish (Felix-Brasdefer, 2005a); iberian and uruguayan spanish (Márquez-Reiter, 2002); iberian spanish and british english (Ballesteros Martín, 2001, 2002;Lorenzo-Dus & Bou-Franch, 2003;) among others.Even though empirical and contrastive studies have been made on different speech acts in a large number of languages since the 1980's along with a vast scientific production, comparative works of that sort are still rare between brazilian portuguese and variants of spanish (Godoy, 2007;Schalkoski-Dias, 2010).
Requests, orders and politeness Austin (1962) and Searle (1969), in the perspective of speech act theory, categorize speech acts based on the functions ascribed to them.Thus, requests and orders are labeled by Austin (1962) as exercitive acts and as directive acts by Searle (1969).In the perspective of both authors, these two acts differ mainly by the presence of the performative verbs (action and order).However, Searle suggests that there may be other illocutionary force indicators.In this work, we adopt the distinction proposed by Sbisà (2009), who considers, on the one hand, the 'request' as a 'behavitive act' since the speaker assigns a 'power' to the addressee.Although, like Austin (1962), she labels the order as an exercitive act, she also extends its characterization in terms of the speaker's power in assigning a duty to the hearer.
Requests are important speech acts: they take place frequently in everyday situations and the desired purpose of the request utterance can implicate a large number of different actions.Taking into account that speakers find in their languages several means adequate for communicating the same meanings, and that these means vary according to the rules of politeness in the speaker's community/culture, theoretical elaborations around politeness have offered a way of explaining some specific linguistic behaviors.The Brown and Levinson (1987) approach considers verbal politeness as a phenomenon centered in the metaphorical notion of face, first formulated by Goffman (1967).For Brown and Levinson (1987), this public face unfolds in a 'positive face' -the desire that every human being has of being looked up to and approved by the others -and a 'negative face', understood as the desire to have freedom of action, and of not being constrained by imposition.It is assumed that, on the one hand, everybody wants to keep their face safeguarded and, on the other, for social relationships to work it is necessary to protect the others' faces.Nonetheless, there are many speech acts which create conflicts of interest and which threaten the speaker's (S) and/or the hearer's (H) face, given that, according to these authors, they go against that which would be desirable to keep the face in a given situation and community.These 'face-threatening acts' (FTAs) can threat both the negative and the positive faces of one or both participants in a verbal exchange.In this sense, requests and orders are considered as speech acts that put the faces of the interlocutors at risk, since they interfere with the freedom of action of H, threatening some needs of her/his negative face.In the Brown and Levinson (1987) approach, given that requestive acts are face-threatening for the hearer because they create pressure on she/he to either perform or not perform an act and thus threaten her/his self-esteem and they also implicate high social risks for both interlocutors, requests need mitigation to compensate for this imposing effect on the hearer.Contrary to Brown and Levinson's view of orders and requests as inherently facethreatening, it is noteworthy that Spencer-Oatey (2008, p. 19, grifo do autor), in her proposal of rapport management, argues that "[…] they 'may' be face-threatening, but need not always be".
Given the need to make a request or give an order, the speaker can minimize the imposition by choosing an indirect strategy over to a direct one.In addition to this, even within a given strategy, there is a diversity of verbal resources available with which to control the degree of imposition involved (Blum-Kulka & Olshtain, 1984).Given that such acts can constitute a threat to, above all, the H's negative face, the languages offer a variety of procedures that are linguistic politeness strategies to which the speakers can resort to mitigate the impact of their requests and which, therefore, enhance their chances of success (Brown & Levinson, 1987).Positive politeness strategies are concerned with safeguarding the positive face of the participants.On the other hand, negative politeness strategies seek to minimize the damage to the hearer's or speaker's negative face.Still according to this perspective, the decisions of the speakers about the verbal strategies to be employed to express a given speech act are related to three social variables: the relative power [P] between S and H; the distance [D] existing between them and the degree of risk or imposition [R] of the act.
In the next sections, we present the categories of analysis of the requests that are usually considered in cross-cultural studies (Blum-Kulka et al., 1989a).

Request's categories of analysis
A request often consists of several sequences of utterances that may precede or follow the main act.In this sense, this may be considered as a discursive macrostructure (Haverkate, 1994).Each of these sequences has a specific function, as shown by the following example, adapted from Blum-Kulka, House and Kasper (1989b) The 'head act' is the part of the sequence that accomplishes the speech act independently of other elements.Moreover, in the classification of Blum-Kulka et al. (1989b), there is a distinction between internal and external modifications of the head act.While the 'external modifiers' refer to the use of 'supportive moves' that modify the impact of the request, mitigating or strengthening the utterance, the internal ones include lexical and syntactical elements that may contribute to reduce the degree of imposition of requests and orders.The categories of external supportive or mitigating moves traditionally described and analyzed in the literature (Blum-Kulka et al., 1989a) are: preparators, getting a precommitment, grounders, disarmers, promise of reward and imposition minimizer.
In Blum-Kulka et al. (1989b), the strategies of external modification of requests are divided in two main categories: the 'mitigating units' and the 'aggravating units'.In accordance with the findings of Ballesteros Martín (1999), in our data we also found occurrences of sequences with neither attenuation nor strengthening of the illocutionary force of the head act, working only to direct the discourse towards the content of what one wishes to request.However, such guiding units, as well as the aggravating ones, were not considered in the present study, since our focus is on the analysis of discursive units that work as external modifiers to the main requests/orders in brazilian portuguese (BP) and in uruguayan spanish (US), bringing a mitigating effect to those acts.
In section 'Results and discussion' we analyze such units, seeking to verify their suitability to our data.

Data collection and procedure
The data analyzed are based on the answers of Written Discourse-Completion Test (WDCT) refined by the authors on the basis of the technique originally developed by Blum-Kulka (1982), and later adapted within the context of the CCSARP -The Cross-Cultural Speech Act Realization Project, (Blum-Kulka et al., 1989a).Since then, the WDCT has been widely used in studies about the production of speech acts in several languages.It consists of introducing the subjects to a small contextual setting from which they are asked to fill in the wording she or he supposedly would employ in a natural conversation.Originally, the WDCT allowed some rule-delimited spacing for the omitted request, followed by a response.However, since Rose found that "[…] the inclusion of hearer response did not have a significant effect on requests elicited […]'' (Rose, 1992, p. 49), questioning the need of including the hearer's answer, since in real communicative exchanges one usually does not know, beforehand, the hearer's reaction, a number of other studies that followed eliminated the rejoinder and used an open questionnaire (Felix-Brasdefer, 2003;Marti, 2006;Economidou-Kogetsidis, 2009;Woodfield & Economidou-Kogetsidis, 2010;Schalkoski-Dias, 2010;among others).Therefore, we have opted to omit the hearer's answer in our WDCTs.
It has been claimed that the amount of contextualized information included in the description of the situation could affect the validity of the results and that detailed contextual information about the setting, time, place, and the kind of relationship of the interlocutors is desirable (Economidou-Kogetsidis, 2012).Taking into account such arguments, we have also broadened the space for the directive act as well as the data regarding the proposed situations, so as to emphasize the variables taken into account in the study and to avoid doubts regarding the illocutionary act that should have been enacted, as shown in the following example.
Acta Scientiarum.Language and Culture, v. 40, e36434, 2018 Situation: You have been a trainee in a medium-sized company for six months and have had little contact with your boss.This month you need some cash advancement.You are still five days away from pay day: your only option is to ask your boss for some money.

You say:
As regards the discursive orientation for the omitted turn, we consider the second person grammatical pronoun used as an informal treatment você / vos (you) more adequate than the third person pronoun (she/he), usually employed in the CCSARP, as it facilitates the adoption of a social role by the participant and the production of the directive act, as she/he imagine how they would behave verbally in the situation, favoring the approximation between speaker and his/her utterance.Indeed, the contributions by the questionnaires' subjects resulted quite close to spontaneous speech, which we consider to have been favored by the modifications we made in the WDCT.
In spite of some limitations attributed to this approach, such as the impossibility of observing interactive aspects of the conversation, (Lorenzo-Duz, 2001;Félix-Brasdefer, 2005b), the WDCT has been shown to be effective in contrastive studies, since it allows the gathering of a large amount of parallel data in relatively controlled situations.Yet the data are not to be seen as displays of authentic discourse, but rather as representative of the perceptions and beliefs of the speakers on the adequacy of the discourse in given situations.Considering the intuitive nature of the data obtained by means of the DCT, we agree with Kasper (2008), who maintains that pragmatic intuition can be a legitimate object of investigation.On the other hand, in view of the argument that, because it is not a really 'natural' situation, the DCT could favor the appearance of answers with stereotypical forms of the required speech act, we concur with Ballesteros Martín (2001) in his consideration that this apparent disadvantage can be seen in a positive perspective, since it "[…] is precisely the more stereotyped or prototypical aspect of the discursive behavior of the subjects, what, in fact, is necessary to know in order to establish significant comparisons" (Ballesteros Martín, 2001, p. 181-182).
In that sense, we believe that the data obtained from the WDCTs can contribute to the understanding of the patterns that guide real communicative exchanges.This is corroborated by the results of the Economidou-Kogetsidis (2013).Comparing WDCT requests and naturally occurring requests, she noted "[…] significant differences in a number of dimensions but at the same time they followed similar trends in terms of directness and lexical modification" (Economidou-Kogetsidis, 2013, p. 21).Following these results, the author argues that "[…] the WDCT requests can, to a certain extent, approximate natural data and that WDCT data is therefore not without validity" (Economidou-Kogetsidis, 2013, p. 21).More recently, Ogiermann (2017) reviews studies comparing DCTs to other data elicitation methods and to naturally occurring data.She considers that while the different data collection methods produce similar speech act realization strategies, the reported differences are mostly inconclusive, with the results depending on the speech acts and groups of speakers under study.

Participants and variables
Data collection involved the participation of 72 brazilian subjects from the City of Curitiba (State of Paraná, Southern Brazil) and 56 uruguayans from Montevideo.All of them are undergraduate students of both sexes, with the majority having ages between 17 and 25.The subjects formulated a request or an order for each of 14 situations presented in the questionnaire, resulting in 892 utterances in portuguese and 722 in spanish.A characterization of situations and variables comprising our WDCT follows: Situation 1 -TV: Husband/wife asks that her/his spouse lower the TV sound volume.'Type of act' -All the 14 situations of DCT have been elaborated in a way that the request implied in the completion of a concrete action by the addressee, such as to lower the TV volume, to give some documents back, to clean the kitchen, and so on.Moreover, considering speech acts as heterogeneous categorical spaces (Hernández Sacristán, 1999), we divide requests/orders into two types: with disapproval (WD) and without disapproval (WTD).This is due to the assumption that, according to contextual configuration, some acts are more threatening to the hearer than others.Therefore, we propose the following characterization to differentiate these two kinds of request/order: a) WD -the request intends to get the interlocutor to change a certain state of affairs -in which he/she is involved -that somehow violates rules previously established and/or which are disagreeable to the speaker (but which is suitable for the hearer).This kind of act would be motivated by some form of 'transgression' by the hearer such as: 'Excuse me, but it is forbidden to park here' (Q38S6); b) WTD -This is the prototypical request/order, that is, one that seeks to lead the hearer to perform a future action, intended to fulfill some need of the speaker: 'Boss, would you help me out and give some cash in advance?' (Q43S9).
'Type of power relationship' -Kerbrat-Orecchioni (2006, p.70) notes that, in some kinds of interaction, "[…] the inequality of the participants is, firstly, a matter of context […]", which includes factors such as age, gender, interactional role, mastery of the language, among others.According to the various power relationships considered in the WDCT situations, the speaker may occupy the following positions: [=P 1 ] addresses a familiar/intimate interlocutor supposedly in equality of power (spouses); [+P 1 ] addresses a close relative, but has legal power over the hearer (parentschildren); [-P] addresses someone in a superior hierarchical position (employee-boss, studentteacher); [=P] has as much power as the interlocutor (friends, colleagues, strangers); [+P] is in a superior hierarchical position (boss-employee); [+P 2 ] addresses a stranger who is in a lower position (passenger-taxi driver, police officer -driver).
'Social distance' -In the elaboration of our WDCT, we did take into account the principle of analysis of conversation according to which […] all interactions take place in a given framework and set some people together, who possess some particular characteristics and some form of socioaffective bond (Kerbrat-Orecchioni, 2006, p. 63).
In this way, for the purposes of the present study, and based on the scheme proposed by Kerbrat-Orecchioni (2006), it will be accepted that the contextual factors that most strongly influence the perception of the horizontal relation are: a) the fact that the interlocutors know each other very well, well, little, or not at all; b) the kind of socio-affective relationship that they have, such as kinship, professional roles, etc.; and c) the nature of the communicative situation, which may be informal or formal.Traditionally, the intercultural studies have considered social distance as a variable with two extremes: [-D], when the distance between the interlocutors is small, and [+D], when there is little or no previous contact between them.Ballesteros Martín (2001), for example, considers [-D] both the situation in which a student requests additional bibliography from a not-very-accessible teacher, and that in which a mother asks the son to go buy something.It appears to us problematic to take these two situations as both having the same level of distance, given that the former refers to an institutional/academic relationship, whereas the latter is a close family relationship.Therefore, in our WDCT version, we have added an intermediate level of distance, [±D], with the following characterization: [-D] -situation in which the people know each other and have a close relationship (kinship and friends); [±D] -when there is no such close familiarity between the participants (professional or academic relationships) and [+D] -when the speaker interacts with a stranger.
'Type of relationship' -The inclusion of the two kinds of relationships (public sphere: professional and academic relationships and contacts with strangers in public places; private sphere: relationships between kinship and friends) as one of the variables is due to the observation that, in general, the studies about the issuing of requests have privileged those situations in the public sphere, mostly institutional ones (Blum-Kulka & House, 1989;Márquez-Reiter, 2000;Marti, 2006;among others).

Results and discussion
As listed in Tables 1 and 2, the most common external modifiers in BP and in US were the 'justifications' (grounder), 'disarmers' and 'thanks in advance'.

Justifications
The data in tables 1 and 2 show that 'justifications' are the most used strategy, both in portuguese (63.19%) and in spanish (56.35%).Except for S8 (bread), our informers used them in all situations, with the highest frequencies in S6 (car) and S9 (payment), both situations in which there is social distance between S and H.It is highly likely that the large number of justifications in S6 is due to the fact that, prior to asking the driver to move his/her car from where it is parked, the police officer explains the motive, that is to say, that it is not allowed to park there.On the other hand, in S9, the request for advance money before the date seems to require a series of manoeuvers by the speaker, as shown in examples 1 and 2, in which the head act is accompanied by several external discursive strategies, among them the justifications: The 'justifications' are units that may appear before or after the head act, providing reasons to give a basis to the S's request.According to Brown and Levinson (1987), to ask or to give reasons for undertaking an action is a strategy of positive politeness, given that it implies that S and H can help each other and, consequently, presupposes cooperation.However, Márquez-Reiter (2000) hold that, by giving reasons, the speaker is also showing appreciation for his addressee, because he or she 2 The examples analyzed were transcribed with as much fidelity as possible to the informers' answers, which, as expected, are close to oral language.
expresses the reason why asking H to interrupt his or her activities and to help S is relevant enough.Therefore, to give and to ask for reasons may be related to both positive and negative politeness.In this way, the high rate of occurrence of justifications in the answers by the participants seems to reinforce this argument of Márquez-Reiter (2000).

Disarmers
The 'disarmers' were the second most used category in both variants (BP: 13.27%; US: 21.23%).They are any form of external modification that prevents the addressee to deny the request.They may include usual promises (I promise this is the last time …), complimentary expressions (I really like your work, but …) and utterances that express the awareness and concern of S that the request might be considered an imposition on H: (3) Maria sei que não é a sua função mas você pode fazer o favor de trocar o dinheiro pra mim? (BP-Q2S12) [Maria, I know it is not your job, but could you change some money for me?] (4) Sé que no es lo que está establecido pero no me queda otra opción que hacer el intento de pedirle un adelanto … (US-Q87S9) [I know I don't have the right to, but I have no option except to ask you some payment in advance] Besides the aforementioned strategies, we consider that some formal markers of politeness, such as com licença [excuse me] and desculpa/e [I´m sorry], in BP, disculpá/e [I´m sorry], perdoná/perdón [forgiveness] and permiso [excuse me], in US, can also work as disarmers to the extent that they express an awareness of S that the request can represent an invasion of the H's territory: (5) Com licença, desculpe o incômodo, mas estou com uma dificuldade financeira, há possibilidade de adiantar um vale?(BP-Q63S9) [Excuse me, sorry to bother you, but I am into some financial problems, could I get some money in advance?] (6) Disculpá, te molesta si te pido que apagues el cigarro?(US-Q121S7) [Sorry, but do you mind to put down your cigarette?]It should be remarked that the apologies found in the previous examples do not work in the same way of a prototypical excuse, which is usually triggered by some previous success on the part of S and which, somehow, did offend H (Searle, 1969).In ( 5) and ( 6), one observes that the apologies anticipate potential impositions at the same time as they announce them.
Traditionally, the formal marker for an apology (and its variations), by preceding the head act, has been considered as an 'alerter' destined to draw H's attention (Blum-Kulka et al., 1989a;Lorenzo-Dus & Bou-Franch, 2003).However, having in mind the examples ( 5) and ( 6), we hold that both the apologetical requests and those asking for permission (com licença, permiso) should not be regarded solely as opening moves for the utterance manifestation of the FTA, but as supportive moves which, by showing deference to the interlocutor, also works as a disarmer.In this sense, we propose to term 'formal disarmer' for such politeness markers.Our proposal to include apologies as a soothing strategy is closer to the perspective of Brown and Levinson (1987), since for them, by apologizing in advance for intending to perform an FTA, the speaker expresses the intention of not imposing on threating the H negative face, partially repairing in advance such imposition.Still according to these authors, besides the formal apologetic requests (I am sorry, but …/ I am sorry to bother you …), S can also communicate his/her intention of not imposing him/herself on H using other strategies that would work as indirect forms of apologizing: a) admitting the imposition (I know you must be very busy, but…); b) showing reluctance (I do not want to bother/interrupt you, but…) and c) giving very strong reasons (I am completely lost…) (Brown & Levinson, 1987).
We assume that, by preceding the imposing act with a request for permission or an apology for the intrusion, S exposes, on the one hand, the vulnerability of his/her positive face.However, on the other hand, he/she also expresses his/her respect for the negative face of H and by doing so reduces the chance that his/her request be denied.
It is well to keep in mind, also, Goffman's notion of 'territories of the self' (Goffman, 1971), related to the self-determination as a coherent and spatiallysituated self.This perspective leads us to the idea of 'territorial offenses' that constitute some incursion on the spatial, temporal or physical territory of the interlocutor (Monteiro, 2008).From this viewpoint, we can say that in utterance (5), the disarmer Com licença (BP) gives evidence of S's awareness that there is a spatial territory that should be respected, since the supposed trainee walks into the boss's office.On the other hand, the formula desculpe o incômodo (5) seems to relate to the concern with the 'temporal territory', given that the attention to the subordinate will require the boss some time.In ( 6), the apologetic request Disculpá (US) seems to refer to another type of territory, which we shall call 'volitive', since the intromission now falls precisely on the wish of action by H, that is, his/her intent of smoking at the place where he/she currently is.
A close analysis of those markers, that we have defined as 'formal markers', revealed significant differences between BP and US, as Tables 3 and 4 show.
The use of disarmers by the uruguayan informers was notably larger than that by brazilians.Besides qualitative differences, there was also variation according to the substrategy preferred in each linguistic variant.Thereby, the marker preferred by brazilians was (com) licença (56%), while in spanish there were only two occurrences of permiso (2.33%), which is the semantically equivalent form.In BP, requests of com licença (excuse me) were used mostly in S9 (payment) and S14 (pen), but it was also resorted to in S6 (car) and S7 (cigarette).Interestingly, these are situations in which there is social distance [+, ±D] between S and H, and which are a part of the public scope.In this case, the variable [P] seems not to have had an impact on the use of the marker, since S9 is [-P], S7 and S14 are [=P], and S6 is [+P].Moreover, in all those situations, the speaker is spatially close to the hearer, which leads us to hypothesize that the use of this marker would be aimed at repairing the incursion into the 'spatial territory' of the hearer.With this device, the speaker would stimulate the latter's receptiveness and willingness, at the same time that he/she would reduce the likelihood of having his positive face damaged by an unfavorable answer.
As regards desculpe/a (sorry) as the second most used marker in BP (43.90%), the predominant situations were also those in which there was social distance between S and H (S6, S7, S9, S13 and S14), with the highest frequency occurring in S13.In these cases, the act of apologizing seems to have the objective of repairing beforehand the intrusion into the hearer's 'volitive territory': in S6 (car), the desire to decide where to park; in S7 (cigarette), the desire of smoking; in S9 (payment), the desire of not being bothered; in S13 (drugstore), the desire to follow the course previously established and in S14 (pen), the desire to keep a certain object.However, in S6 and S9 there could also be a combination of regard for the 'temporal territory' since these are situations in which the fulfilling of a request entails a certain amount of time on the part of the hearer.The uruguayans, on the other hand, have preferred to apologize by using disculpe/á (84.88%), with a larger frequency in S14 (pen), followed by S13 (drugstore), S9 (payment), S7 (cigarette), S6 (car) and S2 (analysis), respectively.Let us remember that the brazilians preferred the (Com) licença requests in S14 and S9, which indicates an intercultural variation in the use of these formal markers in identical contexts.The second formal disarmer most used by the uruguayans was the forgiveness request variant perdone/á (12.79%), alternating with disculpe/á in the same situational contexts.
In summary, given the same contexts, we have found differences in the use of (com) licença and permiso in the two languages.While the brazilian respondents preferred to ask permission to make the FTA, the uruguayans employed, almost exclusively, the forms that refer to an apology request which, in its turn, precedes and anticipates the FTA.Thus, while it may be argued that these markers are semantically equivalent, there is evidence that the com licença (BP) and permiso (US) have different pragmatic values in the two cultural groups, according to our data.Consequently, a discussion is in order on the reach of this contrastive, pragmaticlinguistic study and its possible implications on the teaching of both languages and their variants as foreign languages.To give an example, in the teaching of spanish to brazilian students, it is often stated that permiso is equivalent to com licença, and vice-versa.However, as shown by the pragmatic intuition of the respondents, while the brazilians (from Curitiba) would rather use com licença to approach someone with whom they have little or no familiarity, the uruguayans (from Montevideo) would employ disculpe/á or perdone/á in the same situation.If so, we could say that it is not enough to teach the student that com licença and permiso are semantically equivalent, since, as shown here, the use or pragmatic value of these two formal markers may not be the same (in the two variants) in some contexts.
Considering the disarmers in general, and not only the formal markers, the fact that the occurrence of this strategy is almost 7% larger in the uruguayan corpus (Tables 1 and 2) leads us to conjecture that there would be more concern, on the part of the uruguayans, in safeguarding the spatial territories of the addressee and, therefore, to express more negative politeness, in the terms of Brown and Levinson (1987).Thus, we believe that the more common resort to this type of strategy can be indicative of lesser presupposition (or larger pessimism) that the hearer will fulfill the required action, which is a way to be negatively polite as well.Finally, the predominance of disarmers in S9 (payment), S14 (pen), S13 (drugstore) and S6 (car) in both languages suggests that their use is favored by social distance between S and H, regardless of the kind of request or the kind of power relationship.

Thanks in advance
The 'thanks in advance' were the third modifier most used in both languages (Tables 1 and 2), but its frequency was twice as large in US than in BP.The expression 'thanks in advance' adopted by us refers to the fact that such thanks (Obrigada / Muchas gracias) occur, in our data, without the speaker having received either an affirmative verbal answer or an action by the hearer, as the following examples show: (7) Senhor, peço a gentileza de se retirar desse local, pois aqui como o senhor pode perceber a placa logo a sua frente, é um local de proibido estacionar!Obrigada!(BP-Q11S6) [Sir, please drive away since, as you can see in the sign, this is a no-parking zone!Thank you!] (8) Caballero sería tan amable de mover el coche?Es una zona que debe estar libre.Muchas gracias.(US-Q117S6) [Sir, could you please drive away?This is a no-parking zone.Thank you.] We found no reference to this kind of strategy as a supportive move in previous studies following WDCT, which leads us to suppose that our modifications to WDCT, with the objective of making the answers by the respondents closer to natural language, favored the emergence of that strategy.Traditionally, thanking has been considered as an 'expressive speech act' which gives us information about the speaker's psychological state (Searle, 1976;Haverkate, 1994).However, in the case of the 'thanks in advance' that appears in the responses of our WDCT, we argue that its use in our data may reveal some attitudes related to the a priori assumptions of the speakers.Thus, we surmise that the strategy of giving thanks at the end of requests/orders would play a dual role.On the one hand it appears to attenuate the imposition of the utterance, but on the other, faced with the early thanks, the addressee may feel obligated to comply with the request or order.This means that the speaker, by thanking, is also restricting the options left to the hearer of not performing the acts requested.
In her study of requests in institutional correspondence, Oliveira (1995) makes some remarks that allow one to better understanding the role of thanks as a persuasive move to the overall requesting act.According to her, the illocutionary act of thanking works as an expressive act "[…] whenever it meets the conditions to its accomplishment: someone did something involving a benefit or act of appreciation for the other, who becomes, via the acknowledgements, in debt with the other" (Oliveira, 1995, p. 83) 3 .Still, the thanks "[…] can also work as a directive act, when the debt is taken due to a manipulation of presupposition: the addressee will do as the speaker wishes, hence, the speaker will thank in advance for the performed act" (Oliveira, 1995, p. 83) 4 .
In the view of Brown and Levinson (1987), the 'thanks', as expressive acts, threaten mostly the negative face of S who, while accepting that he/she is in doubt with H, ends up by self-imposing the obligation to be coherent with the feeling that he/she is expressing.On the other hand, from the perspective of Kerbrat-Orecchioni (2006), 'thanks' that work as expressive acts are 'face flattering' to the hearer, being aimed mostly at expressing positive politeness.In the case of 'thanks in advance' in requests and in orders, we assume that they can project a kind of 'pseudo positive politeness', given the directive role mentioned by Oliveira.
Back to the idea of 'manipulation of presupposition' argued by Oliveira (1995), we consider that this notion also applies to our data.However, we have tried to refine it as follows: 'by thanking in advance, S wants H to think that S is certain that H will perform an act A. With this move, S reduces the options by H of not doing A'.
It is therefore possible to relate the view of presupposition manipulation to the "[…] metarepresentational abilities deployed in linguistic communication […]" (Wilson, 2012, p. 330), a notion which refers to the capacity to reason about the 'mental states of others' and that has been particularly important within the literature on theory of mind or mind reading (Horton & Brennan, 2016).With that, we assume that the larger need by the speaker of hedging him/herself with this kind of persuasive resource would be indicative of a lesser certainty of cooperation in a given context, that is, of a lesser degree of certainty that H will act as requested.Therefore, we consider that the larger incidence of this kind of acknowledgement in the uruguayan corpus could also be indication of a lesser presupposition on the part of S that H will be cooperative.The hypothesis is reinforced by the appearance of 'thanks' in the majority of the WDCT situations, which makes it impossible to establish relationships with the variables [P] and [D], or the type of request.

Preparators
The fourth most employed modifier used by the two groups was the 'preparators', but the brazilians used it more often, while the uruguayans used it more in S9 than the brazilians.The speakers employ this discursive strategy to prepare the hearer for what comes next.Usually those moves check the availability (Si tiene un minuto quisiera plantearle una consulta [If you have a minute, I would like to ask you a question] (US-Q116S9) or forward a request of permission (… posso trocar uma palavrinha com o senhor?[May I have a word with you?] (BP-Q3S9) \ Te puedo pedir un favor?[Can I ask you a favor?] (US-Q88S11).Through this procedure, the speaker does not reveal the object of the request and runs the risk of not getting a positive answer.

Securing a previous commitment
'Securing a previous commitment', on the other hand, is not very productive in either variant, with a larger occurrence in S11 (ingredients) and a slightly larger employment in BP.Getting previous commitment has to do with sequences that precede the head act, seeking to engage the commitment of the hearer (…faz uma gentileza para a mamãe?[Can you do Mommy a favor?] (BP-Q39S11)).Although such preliminary moves do not force the hearer to answer positively, the speakers tend to expect positive answers, or silence, with negatives being the least expected answers.Obtaining an early commitment by the hearer makes the speaker feel safer to proceed with the directive act.

Promises of reward
Although with a small frequency, the use of 'promises of reward' was also slightly larger in BP, its use being restricted to private situations [-D], when there is intimacy between S and H (BP: S8, S10 and S11; US: S8 and S11).This is a strategy employed by the speakers to increase the probability of joining the requested act.The hearer will receive the promised reward after the requested action has been accomplished, as in: Prometo que faço sua sobremesa preferida![I promise to cook your favorite dessert!] (BP-Q47S11) / Si vas te quedás con el vuelto![If you go, you can keep the change] (US-Q106S11).

Imposition minimizers
Similarly, the 'imposition minimizers' were not very productive in either language, being most frequent in S8 and S11, which is to say, in contexts of proximity and privacy.As the name suggests, they are procedures that aim at reducing the strength of the directive act, circumscribing its undertaking to specific circumstances so that they do not demand an undue effort by the hearer, as demonstrated in the next example: (9) Amor traga pão para o café.Já que você vai no correio, faz 2 coisas em uma viagem só (BP-Q18S8).
[Love, bring bread for the breakfast.Since you are already going to the post office, you can kill two birds with a stone.]

Conciliatory remarks
Finally, in our data we have also found another type of external modifier, which we shall call 'conciliatory remarks', as found in the sequence underlined in (10).This type of supportive moves was equally rare and used mostly in S5 (list), a situation in which the speaker is in a hierarchically superior position and in which the relationship between boss and employee is somewhat informal [±D]: (10) Carmem está acontecendo alguma coisa?está tudo bem?Eu necessito do relatório que pedi (BP-Q68S5).
[Carmem, is anything going on?Is everything all right?I need the report I asked you.] It can be seen that, before reiterating the previously made request, the speaker is sympathetic with the hearer by means of strategies of positive politeness, as in the case of showing interest in the hearer.Moreover, at the discursive level, to the extent that they leave more room for negotiation as regards the hearer's obligation to fulfill the act, we consider that those strategies may also be expressing negative politeness.However, it must be observed that, precisely in S5, there were also a considerable amount of aggravating unities, such as disapprovals, calling the hearer's responsibility, and threats.This is to say that, when the request/order is laden with an implicit disapproval, the concerns with the face's sensitivity seem to be minimized.
We have not found any reference to this kind of modifier in the literature.At any rate, according to the role that they seem to play in example (10), they seem aimed at preserving harmony, which is to some extent threatened, between the speaker and the hearer.

Final considerations
The studies of Blum-Kulka et al. (1989a) already showed that each culture and language have its preferences for certain kinds of verbal strategies.In fact, the quantitative analysis of our data has confirmed that, although the two languages studied coincided on the most used categories, there was variation in the category frequencies, such as 'justifications', 'disarmers', and 'thanks in advance'.The most frequent occurrence of the latter two in answers by the Uruguayans favors the interpretation that there is a larger tendency to express negative politeness among them, since they tend to be more pessimistic about the fulfillment of their requests.However, the differences found do not mean that one group is more polite than the other, since there are many ways to modify the strength of the head act internally, which we did not discuss in this work.As regards the external mitigating moves, we took the view that, mostly, it is the speakers' perceptions and valuations which are at stake, given the sensitivity to the social faces and above all the type of discursive strategy most suitable to each context.Even though the data analyzed here are not instances of natural speak, we believe that they portray the pragmatic intuition of the respondents, and that they faithfully mirror the most adequate uses in each context as viewed by the respondents.Finally, considering our results on the 'formal modifiers', we wish to emphasize that the rather different use of the expressions com licença and permiso, although semantically equivalent, may well lead the students of spanish/portuguese as a foreign language to a pragmatic mistake, precisely because they are not used in the same contexts in the two variants studied in this work.
[WD, = P 1 , -D, Private] Situation 2 -Assignment: Student asks that teacher return written assignment.[WD, -P, ±D, Public] Situation 3* 1 -Kitchen: Young adult asks her/his housemate to clean their kitchen.[WD, =P, -D, Private] Situation 4 -Backpack: Father/mother demands that teenage daughter puts way backpack and sneakers left in inadequate place.[WD, +P 1 , -D, Private] Situation 5 -List: Boss requests that his secretary bring a list of prices that he had already previously asked for.[WD, +P, ±D, Public] Situation 6* -Car: Police officer asks that a driver move his vehicle from the spot where it is parked.[WD, +P 2 , +D, Public] Situation 7* -Cigarette: Young person asks a stranger to put down the cigarette.[WD, =P 2 , +D, Public] Situation 8 -Bread: Husband/wife asks her/his spouse to buy bread on the way home.[WTD, =P 1 , -D, Private] Situation 9 -Payment: Trainee asks her/his boss an advance payment.[WTD, -P, ±D, Public] Situation 10 -Clothes: Young person asks a friend to lend her/him a change of clothes.[WTD, =P, -D, Private] Situation 11* -Ingredients: Father/mother asks teenager son to buy some ingredients for supper.[+ P, private] [WTD, +P 1 , -D, Private] Situation 12 -Change: Boss asks employee to go out and change some money.[WTD, +P, ±D, Public] Situation 13 -Drugstore: Passenger asks taxi driver to stop at a drugstore which is out of their way.[WTD, +P, +D, Public] Situation 14 -Pen: Young person asks a stranger to lend her/him a pen.[WTD, =P, +D, Public]

( 1 )
[disarmer] Chefe, desculpa incomodar mas [preparator] tô precisando pedir algo.[disarmer] Sei que hoje ainda é dia 15, mas [justifications] preciso muito do vale.Não paguei minha faculdade ainda e não queria deixar os dias passarem para pagar mais juros.[head act] Tem como só este mês adiantar o vale agora? 2 (BP-Q37S9) [[disarmer] Boss, sorry to bother but [preparator] I need to ask you something.[disarmer] I know it is still the middle of the month, but [justifications] I badly need some money.I haven't paid college yet, and I would need to pay interest if do not pay it right away.[head act] Is it possible for you to give me some money in advance?] (2) [disarmer] Me da mucha vergüenza pero [justification] tuve un problema estos días y [head act] necesitaría un vale antes de la fecha.(US-Q127S9) [[disarmer] I am deeply embarrassed but [justification] I have had some problems these days and [head act] I need some money in advance]

Table 1 .
Frequency of supportive moves in brazilian portuguese (BP).

Table 2 .
Frequency of supportive moves in uruguayan spanish (US).

Table 3 .
Distribution of formal disarmers by situation -BP.

Table 4 .
Distribution of formal disarmers by situation -US.