1. INTRODUCTION

According to Bosque (1980), the Spanish past infinitive (independent infinitive + past participle, i.e. haber estudiado) usually expresses an imperative value, as can be seen in the following example, extracted from an online forum. The speaker quotes another participant whose post included a past infinitive, rephrasing this construction as an imperative:

These expressions, also known as retrospective imperatives (Bosque 1980), combine aspectual and deontic information. They refer to actions the addressee performed in the past and which the speaker considers problematic. The use of haber + past participle to indicate an incongruence between the action and what the addressee should have done leads us to believe that the construction is used to convey reproach.

In this study we establish three research questions. First, what are the main attributes that characterize haber + past participle? To this end we aim to identify which features characterize the construction and to provide support for the analysis through the use of examples from corpora. Once these attributes are clearly established, we will assess the extent to which the inherent features of the construction are also observed in the illocutionary act of reproach. In other words, can we observe a high degree of coincidence between these features and the illocutionary act itself (and thus a specialization of the construction), or, on the contrary, does a considerable dependence on the context limit the relation between haber + past participle and the speech act of a reproach? Finally, and related to the previous question, what discursive patterns can be identified in the use of the construction? For such a study, we consider an empirical approach to be optimal, one that allows us to support our conclusions with authentic data, and hence real examples of language use have been extracted from a variety of corpora, towards a better understanding of the ways in which the retrospective imperative emerges in the language.

2. PAST INFINITIVE: IMPERATIVE OR CONDITIONAL?

Only a limited literature on the past infinitive and its characterization in Spanish is available, and there are several possible explanations for this. First, whereas the construction itself is rather common (Bosque, 1980) it is not always easy to document in corpora (Van Olmen, 2017: 117); moreover, in the Hispanic context, it is perhaps not widely or evenly spread across all Spanish varieties³. For our present purposes of characterizing the form, we will draw on the works cited above, as well as on the research of Biezma (2010) and Vicente (2013).

Bosque (1980) was the first study to draw attention to haber + past participle, observing as it did coincidences between the past infinitive and the imperative. First, sentences with the past infinitive do not have a truth value; that is, ‘true/false’ is not a felicitous reply. Another common trait of imperatives is that the past infinitive cannot be embedded. Finally, it can only have a 2^nd-person addressee, in that the past infinitive is always other-oriented. These characteristics lead Bosque to call such infinitives retrospective imperatives, since they share traits with imperatives but, unlike imperatives, refer to past actions.

Biezma (2010) disagrees with both Bosque (1980) and Vicente (2013) as to the imperative nature of this structure, identifying it rather as a hidden counterfactual conditional. The structure haber + participle clause, she argues, is the result of an ellipsis of the apodosis. On these lines she claims that in structures such as (2a) the speaker spells out only the inverted antecedent, which is shown here in (2b).

From this perspective, the segment haber cotizado más! (you should have paid more taxes!) is equivalent to si hubieras cotizados más (if you had paid more taxes) and is inverted because it presupposes that the addressee hasn’t paid the right amount of tax to qualify them for the social security benefit. The apodosis of the conditional is left unsaid in (3a) due to the common ground shared by the participants, which makes it unnecessary. Regarding the absence of the conditional conjunction si, Biezma (2010: 9) points out the existence of similar structures in rural parts of Castile (a central area of Spain) where the si is typically missing but where the construction makes perfect sense.

Both these approaches help us to understand the grammatical, semantic and syntactic complexity of this structure, yet they also have certain limitations. On the one hand, interpreting haber + past participle as an imperative requires a flexible understanding of what exactly an imperative is. For example, imperatives have a present or future-oriented interpretation, while the structures under study here refer to actions that should have been performed in the past.

On the other hand, conditionals are not limited to the 2^nd person, whereas haber + past participle structures are addressed to the interlocutor; we will, however, discuss this issue further in section 4.1.2. Vicente (2013: 40-48) offers various arguments against the interpretation of the construction as an inverted conditional. For example, a conditional reading of haber + past participle ignores the directive value that these structures have (in contrast to conditionals, which can carry a wider variety of illocutive forces), and conditionals do not require a volitional agent, while the haber + past participle does.

The somewhat contradictory traits of these different interpretations and the difficulties to match all the characteristics of the imperatives and conditionals has led to see haber + past participle as a construction of hybrid nature (Van Olmen, 2017). This position allows to integrate in a sense both perspectives, as it understands haber + past participle as a construction of its own⁴.

Lastly, all authors agree that the past infinitive can be used to make the addressee aware that some dispreferred action has been performed. At the same time, the authors of existing studies recognize that their analyzes are based on personal experience and ad hoc examples, which limits the scope for any definitive assessment of the discursive behavior of the past infinitive. The present study, then, seeks to fill this gap in the literature.

3. CORPUS AND METHODS

This study aims to present a discursive characterization of the Spanish construction haber + past participle as a form which is specialized in encoding acts of reproach. One of the novelties of this study, in comparison to existing work in the area, is that it is based on the analysis of various corpora of natural language. The first of these, the ESPRINT corpus (Albelda and Estellés, coords., unpublished), is an oral corpus of spontaneous conflictive interactions between couples who reported having relationship and marital difficulties. In terms of non-conflictive oral corpora, or at least those in which conflict was not a criterion for compilation, we have used the Val.Es.Co. corpora (Briz y Grupo Val.Es.Co., 2002; Pons Bordería, online), the COJEM Corpus (Méndez Guerrero, 2015), the ESLORA Corpus 2.1. (online) and the AMERESCO Corpus (Albelda and Estellés coords., online). Finally, we semi-manually extracted examples from the Corpus del Español: Web/Dialects (Davis, online), which collects written texts drawn from the Internet. This last corpus added depth to the study, in that we were particularly interested in observing whether the construction in question was found in texts written in informal contexts.

In contrast to the other corpora used, the Corpus del Español (see Table 1) was so large that searches for haber yielded too many results to be analyzed. Instead, the sequence of haber followed by certain participles (haber estudiado, haberlo dicho, haber elegido) and the sequence no + haber were considered. In addition, we observed that the construction we were interested in often appeared after a pause (whether a period/full stop or a comma), so this was also taken into account in the search strings.

Lastly, we identified cases of haber + past participle in the corpora that work as independent constructions, that is, they have syntactic autonomy, independent illocutive force, and can constitute a full intervention.

4. RESULTS

A total of 55 examples were retrieved from the corpora, the majority being cases from written texts drawn from the Internet. Of note, most examples were found in the comments sections of blogs or in posts on forums. Despite the written form of such examples, computer-mediated communication has been shown to bear certain similarities with oral conversation (Herring, 2010); for example, there is typically limited planning time, as in speech production, which favors immediacy and spontaneity and hence leads to a greater use of the colloquial register. In this sense, our data suggests that the retrospective infinitive construction in Spanish is marked as being notably present in oral and informal discourse.

Turning to the examples extracted from colloquial conversations, most of the examples were from conversations occurring in harmonic situations (14 cases) rather than broadly conflictive ones (3 cases). At first sight this appears to run contrary to our initial hypothesis, which assumed the reproachative nature of the construction. However, in addressing the data from a qualitative point of view, we observed that some discrepancies arise within harmonic conversations which can explain the presence of haber + past participle. In other cases, contextual parameters may deactivate or mitigate the potential hostility of the construction (see section 4.3.1.).

4.1. Characterization of structure

There are six traits that allow us to characterize this structure further: 1) the need for it to appear in a reactive turn or to be understood as a reaction; 2) its appellative nature, in that it is addressed to a 2^nd person; 3) its present-future orientation although referring to actions performed in the past; 4) its use in expressing counterfactual values; 5) the requirement of agentivity; and 6) its potential to encode a hostile act.

4.1.1. Reactive turn

Haber + past participle cannot be found in discourse-initiating interventions. Unlike imperatives, these constructions require a trigger in order to be uttered. Evidence of this reactive nature is that the construction frequently appears preceded by pues (well, then), as in (4), and indeed in 36.3 % of the examples in our data.

In the above example, two couples are dining together. F says that she was tempted to buy something from a catalogue because the purchase would have earned her points on a reward program, leading to a free trip to Majorca. Her husband (J) discourages F from doing so because he cannot ask for more days off work. M then jokingly says that they should have bought the product anyway and given the trip to her and her husband, using pues + haber + past participle to express this.

Portolés Lázaro (1989: 131) identifies an adversative value in pues: «If a speaker says p and their interlocutor replies pues q, we must think that q contradicts some conclusion that could somehow be inferred about p, orienting the dialogue towards a different one»⁵. In this case, the impossibility of J asking for more time off work leads to a conclusion: F and J acted correctly. However, M humorously reprimands J, and then introduces a new conclusion: it would have been possible to buy the product in order to win the trip to Mallorca and then to ask her and her husband to take the trip instead.

Vicente (2013: 52) has argued that this construction does not always appear in a reactive intervention. He proposes that its reactive character must be understood as a tendency, since it can also be found in an initiative position, as in the following example, which he himself gives.

Strictly speaking, there is not a prior verbal intervention that makes it possible to claim that B’s utterance is reactive, in a similar way to there being no previous intervention when someone says «thank you» to a person who has opened the door for them. However, neither of these two utterances would be possible had the previous action (not passing the ball or opening the door) not been performed. Some activities are not structured exclusively by means of verbalization, but are a combination of the actions that participants take and the way in which these relate to objects and the environment (Evans, 2017). In this sense, these examples can indeed be understood as reactions to previous actions, regardless of whether that action is verbal or non-verbal.

Furthermore, due to the eclectic nature of the corpora used in this study, we have found several such examples in written sources. Even in these cases, the reactive value can be confirmed, in that the speakers strive to recall possible previous interventions to which they respond with haber + past participle.

4.1.2. Prototypically addressed to a 2^nd person

Given its directive nature, the structure is usually addressed to the person to whom the 2^nd speaker is talking (see examples 3 and 4 above). However, Biezma (2010: 5) provides an example where the referent seems to be a non-present 3^rd person. We have found similar examples in the corpora, as (6).

From a grammatical point of view, this case is a reaction to a non-present 3^rd person where haber + past participle allows the speaker to show dissatisfaction with the attitude or conduct of that 3^rd person. However, from a discursive perspective, such cases can also be understood as a reframing of the situation, with the speakers expressing what they might have said to the addressee had they been present. The reference to a prior discourse or action performed by the professors allows for a shift in the discursive parameters, bringing them closer to the speaker at an enunciative level. In this sense, the infinitive is uttered as if the addressee were indeed present. Moreover, this example includes an element which endorses the use of haber + past participle as part of direct speech, such as the presence of the reactive and counter argumentative discourse marker pues. However, this marker is not always present, especially in oral discourse, where direct speech can be conveyed through the use of prosodic features (Estellés Arguedas, 2015).

In other cases, this polyphonic interpretation arises though the use of quotative devices, such as the quotative que or conditional clauses (7). Examples such as the following represent 13% of the occurrences in our corpora.

The retrospective imperative constitutes the apodosis, while the protasis includes the reported speech or situation to which the haber + past participle is a reaction. In terms of our collection criteria, these infinitives are not syntactically independent, but rather interdependent⁶. Another example (ad hoc in this case) is provided by Vicente (2013); in contrast to the previous examples, there is a 3^rd-person singular pronoun, which makes it difficult to apply an interpretation involving reframing.

We did not find examples like this in the present corpora. As an alternative, we informally asked native peninsular Spanish speakers – some of whom had studied Spanish Linguistics and others not – if they would say or had heard the structure presented in (8). We found some variation as to the degree of acceptance within these two groups of informants, which might well be seen as a sign that the prototypical use of haber + past participle is to address a 2^nd person. Nevertheless, more research needs to be done towards establishing stronger arguments in support of such a claim.

4.1.3. Referring to the past but future-oriented

Linguistic devices similar to the past infinitive can be found in other languages, such as Dutch (Duinhoven, 1995; Van Olmen, 2017), where a verbal form of the past is used to talk about what should have happened. In Spanish, despite the reference to the past being marked through the aspect of the past participle, in 25% of the cases that we analyzed this construction is followed by adverbs and temporal phrases that highlight this reference to the past, such as antes, a su debido tiempo, cuando tocaba, cuando estabas a tiempo (before, in due time, when it was time, when you still have time), among others.

This feature entails a paradox in that the speaker uses a past form to command or admonish, yet these actions are present or future-oriented. In such cases, the future orientation is not conceptualized as something the addressee should do in the future, but something they should have done to achieve a different outcome in the future of that past. It also prompts the addressee to reflect on what they have done and invites them to acknowledge the error of their actions and to recognize the consequences (Van Olmen, 2017).

4.1.4. Expressing counterfactual values

The structure of the enunciation, uttered in the present, in fact refers to a point in the past where a dispreferred action has taken place «and then returns (to) a set of possible worlds lying on the future of that past point» (Vicente, 2013: 15). To illustrate this counterfactual value, we will rephrase example (3a) as if A had paid her taxes.

Formulated in this way, B’s intervention is pragmatically odd, since A has acted as expected in the past to get her medicine free through the social security system. In other words, the counterfactual value has been deactivated and thus the use of haber + past participle does not work.

4.1.5. The requirement of agentivity

Regarding the directive value of this structure, it is necessary that the addressee to whom the message is addressed has the capacity to carry out the required action. If we consider example (7) again, we can see that the speaker’s utterance is felicitous because the parents, in the past, were in a position to choose whether or not to stay with their child, or this is at least what the speaker believes to be true.

Although less frequent in our data, we have retrieved similar examples where the power of the addressee to perform an action, and thus to be held responsible for the undesired present situation that they are experiencing, is limited. This can be seen in (10) with the verb nacer (‘to be born’). A family is talking about an appropriate time for A to leave the family home. B, an older relative of A’s, has previously stated that he bought his flat when he was 21 years old and explained the effort involved in leaving his parents’ house. A then explains that nowadays it is difficult for someone in his position to become independent in this way, adding that his relatives can achieve this independence for him by giving him an apartment.

During the interaction, A playfully claims that the parents of an acquaintance (Ainhoa) bought her a flat, implying that to give someone property as a gift is not so extraordinary, and maybe the interlocutors here could do the same for him. An older relative (C) replies that in order to get a flat for free he should have been born as the son of Ainhoa’s mother (Lidia). Clearly, A’s ability to have performed such an action in the past is nil, hence a humorous interpretation is activated.

4.1.6. Potential hostility

It seems that retrospective imperatives are prone to display a certain degree of hostility. In fact, Biezma (2010: 7) qualifies them as «pretty rude» in that they carry a sense of obviousness. Besides the fact that making someone aware of the error in their actions is a potential threat to the addressee’s face, from a discursive point of view we can also perceive elements that signal potential hostility. In our data, we have observed that swear words, pejorative address forms and appellative forms often occur adjacent to the past infinitive, as well as face-threatening acts expressing counterfactivity.

Does this mean that the use of this structure necessarily implies rudeness? The answer is no, but there is a tendency to display some hostility. The use of haber + past participle can be found in situations where speakers do not use it in a negative way, as illustrated in (11).

The first speaker discovers that his interlocutor has found a job and wants to celebrate it. In this context, haberlo dicho/you should have said it can hardly be considered rude, in that the speaker’s intention to celebrate the addressee’s good news is clear. However, what is inherent to the construction (no matter if it is in a harmonic interaction or oriented to a good outcome) is that the speaker who utters the past infinitive shows some dissatisfaction about the addressee’s actions in the past, and in doing so makes the addressee responsible for his actions that led to the present undesired situation. According to Vicente (2013: 19), the degree of perceived rudeness or hostility will depend on various factors, including the meaning of the retrospective imperative, the extent to which the speaker is dissatisfied, and the context, among others.

4.2. Retrospective imperative: a form of reproach?

So far, we have characterized the construction haber + past participle based on empirical data drawn from a number of corpora. We will now turn to our second research question, the aim of which is to determine whether these infinitives behave like a specialized form of reproach. It should be noted that some authors (Biezma, 2010; Bosque, 1980; Van Olmen, 2017; Vicente, 2013) have claimed that the construction is indeed used to express a reproach. However, we believe that some reflection is needed regarding the traits that have been cited as defining the (claimed) reproach here, as well as on the way that such reproaches are realized in real examples. Let us take the following example, from a conversation between a group of friends talking about the purchase of a computer, in which haber + past participle is used as a reproach.

First, prior to making a reproach, a speaker must perceive a contradiction in the interlocutor’s words, actions or thoughts (Burguera Serra, 2010; Plantin, 2005). Once the contradiction is identified, the speaker reacts to it by expressing their disapproval. A discursive consequence of this sequence of actions is that reproaches are not only reactive (they require previous activity, whether verbal or non-verbal) but are also usually initiative; when a reproach is uttered, the speaker expects a reaction, preferably a reflection on the addressee’s rights and duties and/or a change in the present or future course of action. For this reason reproaches are situated on the past-present or past-future axis (Burguera Serra, 2010: 405; Van Olmen, 2017). In the example here, B notes a contradiction in A’s behavior: A has purchased a Macintosh instead of a PC because he is not a computer expert; from the perspective of B (an expert) a PC would have been the better option. The actions discussed occurred in the past, but have consequences in the present (A has not chosen wisely).

Reproaches are addressee-oriented, irrespective of whether the addressee is present or not. The addressee is being judged and/or criticized by the speaker, who considers that their interlocutor is responsible for the present bad situation and seeks to provoke a reaction. This reactive-initiative nature, plus the feeling of being judged, leads to reproaches tending to escalate into verbal conflict (Albelda Marco, 2022: 23). As a consequence, reproaches threaten the face of participants and are usually categorized as instances of impoliteness (Burguera Serra, 2010; Ilie, 1994). In example (12), B addresses A aggressively, judging him to have been responsible for a bad choice (buying a Macintosh instead of a PC). A senses this attack and feels the need to reply with a justification of his actions.

In terms of participants’ rights, when a reproach is uttered, the speaker feels morally and deontically entitled to condemn an action and to incite the interlocutor to acknowledge their error and either to emend the error or to recognize it as such. Thus, the speaker projects a deontic and socio-functional asymmetry on the addressee (Stevanovic, 2018; Stivers, 2008); an epistemic asymmetry (García Ramón, 2018; Heritage, 2012) also arises, in that the reproach is conceived as a reminder of the addressee’s duty to act in the right way. Returning to our example, B not only feels entitled to criticize A’s purchase, but also presents himself as expert in the matter, as can be inferred from B’s last intervention.

Finally, reproaches are eminently directive. However, like speech acts such as apologies and justifications, they also present features of expressive illocutionary force, and hence can be understood as acts with a hybrid illocutive force in which the directive and the expressive force combine (Albelda Marco, 2022). The expressive value stems from the negative judgement that a speaker makes about an addressee’s past actions (like buying a Macintosh instead of a PC). From the speaker’s point of view, an expectation of what should have been done is not met, and this grants them the right to express their dissatisfaction. As for the directive value, we have already described the desire to provoke a reaction in the interlocutor.

From what we have seen, then, the characteristics of the construction and the traits that define the act of reproaching have a lot in common. Hence it seems safe to say that this construction has a high degree of specialization in conveying reproach. However, our data also reveal that it is not every case in which haber + past participle appears in a statement that it can be interpreted as reproachful. Likewise, since reproaches are acts of a reactive-initiative nature, it is also necessary to analyze replies to these statements to better understand the way in which the construction is used, and how it can be interpreted from a discursive point of view.

4.3. Discursive patterns

In this section we will address the issue of discursive patterns, understood as a «recurring practice in the configuration of discourse that, without reaching a fixed form, constitutes a frequent routine when arranging informative materials»⁷ (Taranilla, 2015: 260). To this end, attention must be paid not only to the form itself and the linguistic devices and structures that usually appear with the past infinitive, as discussed above in section 4.1., but also to the way in which the infinitive is used in the construction of discourse and in successive interventions. From this perspective, it will be possible to identify patterns of usage, as summarized in Figure 1, which we will address in more detail in what follows.

Figure 1. Discursive patterns in harmonic and conflictive interactions

4.3.1. Neutralization and mitigation of the reproach

The neutralization of a reproach occurs when some of the characteristics described in the previous sections are not present. For example, in (16) speaker C recriminates her younger relative for not having been born to a different person (haber nacido de Lidia/ you should have been born to Lidia). Since the requirement of agentivity is not met, the addressee cannot be held responsible for their past actions and thus the reproach is neutralized.

In other cases, all the characteristics are present, but there may also be other elements (such as ironic and humorous values or pseudo-impoliteness) that lead to a less menacing interpretation. Consequently, the reproach can be mitigated or even neutralized, as illustrated in example (4), here shortened and renumbered as (13).

In (19), M’s intervention may seem to be a reproach, in that she scolds her friend for not having made a purchase that would have resulted in getting a free trip to Majorca which in turn she could have given to the speaker (M) and her husband. She even brands her interlocutor as selfish for not having done so. This exaggeration, and the fact that M’s friends are not responsible for organizing M’s holidays, triggers a rather humorous or ironic interpretation of the utterance. However, can we be sure that M is in fact satisfied with her friend’s decision, and that she does not expect to be consulted if a similar opportunity arises in the future? In other words, is she appealing for the reconsideration of a past action in order to encourage a reflection on it, or to encourage a different way of acting in the future? In the following interventions M continues with the joke until J explains that the item they had to buy was too expensive. It seems that in such cases it is not absolutely clear that the reproach is neutralized, but the harmonic context favors a diluted and playful interpretation.

4.3.2. Rebuttal and disagreement

When haber + past participle is uttered in conflictive dialogic interactions, the addressee can respond in two ways. First, they might agree with the speaker that the action in the past was not performed and try to justify their actions and make amends in some way. This is what happens in (14), where a couple interact while driving a car.

In this example, a woman reproaches her husband for not having calculated that they are not able to stop at a service area on the highway because there are no shops or picnic areas. The husband’s reply serves as a defensive strategy, agreeing with his wife (no nos paramos/we didn’t stop) and justifying himself, as he is now actively looking for a good picnic spot for her.

The following example illustrates a case of a counterattack. A couple are discussing picking up their children from school by car.

The infinitive is used here by the wife (Haber cogido los coches, guapo/You should have taken the cars, darling!). In the following intervention, the husband replies using disagreement, which can be understood as another reproach: I take the kids every day and I pick them up the rest, it’s my daily routine, uh? In our data, this value of disagreement and counterattack is the most frequent in conflictive conversations.

4.3.3. Argumentation

In monologic contexts, we can find cases where the infinitive appears directed to a non-present interlocutor that the speaker invokes or recalls from a past or hypothetical intervention or situation. The interlocutor is not present, and thus cannot make an immediate response, and this allows the speaker to develop their line of argument. In our data, we have observed three ways that this argumentation develops: through the addition of further reproaches or FTAs, an extension of the reproach, and the expression of beneficial consequences of the advice given. Example (16) illustrates the first of these.

Despite the apologies and the affectionate forms of address, new reproaches compound and magnify the (hypothetical) reprimand that the speaker is making to his not-so-beloved wife for not taking good care of her appearance, among other things. As a result, the threat to the addressee’s face is heightened by the accumulation of reproaches.

Another way of developing the argument is to extend a reproach. In the following example, the speaker is posting on a forum a response to someone who complained in a previous post about spending too much time taking care of their child.

As in the previous example, the use of an apology before the reproach is insincere since what follows conveys no empathy for the parent’s situation. Moreover, the next utterances are an expansion of what the speaker has implied (haberlo pensado antes/you should have thought it before) because it presents the behavior of the child as something normal that the parent should have expected (and therefore accepted) when they decided to have a child.

The final possible pattern differs from the previous ones in that it makes explicit the positive consequences that would have been achieved if the addressee had acted differently in the past.

The interlocutor is reproached for not having studied, because had he done so he would now have more leisure time and a better quality of life. After uttering the infinitive, the beneficial consequences of having followed the advice conveyed by the reproach are expressed.

5. CONCLUSION

This study has explored the ways in which the perfect infinitive is discursively configured when used as an independent construction (or an interdependent one, when occurring within a conditional clause). To this end, we have used data from a variety of corpora, both oral and written in nature, focusing on both harmonious and conflictive situations. The analysis reveals that this construction has a marked oral and colloquial character, something which helps to explain the difficulty that some authors have noted in addressing the phenomenon.

Our analysis has revealed six attributes linked to haber + infinitive: it typically constitutes a reaction to a previous action, thought or utterance; prototypically it appeals to a 2^nd person; it refers to the past but is present and future-oriented; it expresses counterfactual values; it has the requirement of agentivity; and it is prone to convey hostility. If these attributes are compared to the criteria that characterize reproaches in general, we observe that there is a great deal of coincidence. This leads us to conclude that there is a high degree of specialization in this structure when it conveys reproach.

Regarding the construction of discourse, three major discursive patterns have been distinguished. First, it is possible that some of the traits that prototypically characterize the structure are not present (agentivity, for example) or that there is some element in the context that deactivates or mitigates the act of reproach (i.e. irony). Second, if a reproach occurs in a dialogical context, the recipient of the reproach can choose to self-justify (which can de-escalate the conflict) or to counterattack as a way of endorsing the conflict (which is the most frequent case). Third, in monological contexts, reproaches are used as a means of developing the argument. In this sense, it is possible to add new reproaches or face-threatening acts, extend the reproach or, less frequently, note the positive consequences that would have arisen if the interlocutor had acted differently.

REFERENCES

Albelda Marco, Marta (2022). ¿No te das cuenta de que me estás molestando? Las recriminaciones intensificadas mediante preguntas retóricas. Oralia, 25(3), 9-32. https://doi.org/10.25115/oralia.v25i2.8065

Biezma, María (2010). Inverted antecedents in hidden conditionals. En S. Kan, C. Moore-Cantwell, & R. Staub (Eds.), Proceedings of North East Linguistics Society 40. Graduate Linguistics Student Association. https://mariabiezma.files.wordpress.com/2012/10/biezmapaperrevised.pdf

Bosque, Ignacio. (1980). Retrospective Imperatives. Linguistic Inquiry, 11(2), 415-419.

Burguera Serra, Joan Gabriel (2010). Gramática y pragmática de la interrogación retórica en español. Una aplicación al debate parlamentario [Ph.D. Thesis, Universitat de Barcelona]. En TDX (Tesis Doctorals en Xarxa). https://www.tdx.cat/handle/10803/1705

Duinhoven, Antonius Maria (1995). «Had gebeld!» De irreële imperatief. Tijdschrift voor Nederlandse Taal- en Letterkunde, 111(4), 346-364.

Estellés-Arguedas, Maria (2015). Expressing evidentiality through prosody? Prosodic voicing in reported speech in Spanish colloquial conversations. Journal of Pragmatics, 85, 138-154. https://doi.org/10.1016/j.pragma.2015.04.012

Evans, Bryn (2017). Sports coaching as action-in-context: Using ethnomethodological conversation analysis to understand the coaching process. Qualitative Research in Sport, Exercise and Health, 9(1), 111-132. https://doi.org/10.1080/2159676X.2016.1246473

García Ramón, Amparo (2018). Epistemicidad en interacción: (A)simetrías epistémicas en secuencias de acuerdo y su relación con la construcción de roles funcionales en conversaciones y entrevistas [Doctoral thesis]. https://roderic.uv.es/handle/10550/69401

Goldberg, Adele E. (1995). Constructions: A construction grammar approach to argument structure. University of Chicago Press.

Gutiérrez Ordóñez, Salvador (1997). La interdependencia en sintaxis. Cauce: Revista Internacional de Filología, Comunicación y sus Didácticas, 20, 703-725.

Heritage, John (2012). The Epistemic Engine: Sequence Organization and Territories of Knowledge. Research on Language and Social Interaction, 45(1), 30-52. https://doi.org/10.1080/08351813.2012.646685

Herring, Susan C. (2010). Computer-mediated conversation Part I: Introduction and overview. Language@Internet, 7(2). http://www.languageatinternet.org/articles/2010/2801

Ilie, Cornelia (1994). What else can i tell you?: A pragmatic study of english rhetorical questions as discursive and argumentative acts. Almqvist & Wiksell.

Méndez Guerrero, Beatriz (2015). Corpus Oral Juvenil del Español de Mallorca (COJEM). https://ebuah.uah.es/dspace/handle/10017/25298

Plantin, Christian (2005). L’argumentation: Histoire, théories et perspectives. Presses Universitaires.

Portolés Lázaro, José (1989). El conector argumentativo «pues». Dicenda: Estudios de lengua y literatura españolas, 8, 117-134.

Stevanovic, Melissa (2018). Social deontics: A nano-level approach to human power play. Journal for the Theory of Social Behaviour, 48(3), 369-389. https://doi.org/10.1111/jtsb.12175

Stivers, Tanya (2008). Stance, Alignment, and Affiliation During Storytelling: When Nodding Is a Token of Affiliation. Research on Language and Social Interaction, 41(1), 31-57. https://doi.org/10.1080/08351810701691123

Van Olmen, Daniël (2017). Reproachatives and imperatives. Linguistics, 56(1), 115-162. https://doi.org/10.1515/ling-2017-0033

Vicente, Luis (2013). Past counterfactuality in Spanish imperatives and its implications for the analysis of imperatives. https://semanticsarchive.net/Archive/DM5ZGY1N/vicente---past-counterfactuality-in-spanish-imperatives.pdf

CORPORA

Albelda, Marta & Estellés, Maria (Coords.). Corpus Ameresco, Universitat de València, ISSN: 2659-8337, Available online at <www.corpusameresco.com>.

Albelda, Marta & Estellés, Maria (Coords.). Corpus ESPRINT, Universitat de València (unpublished).

Briz, Antonio, & Grupo Val.Es.Co. (2002). Corpus de conversaciones coloquiales. Anejos de la revista Oralia, 1, 1-383.

Davies, Mark (Dir.) El corpus del español (Web/Dialects). Available online at <https://www.corpusdelespanol.org>.

ESLORA: Corpus para el estudio del español oral, 2.1 version june 2022, ISSN: 2444-1430. Available online at <http://eslora.usc.es>

Pons Bordería, Salvador (Dir.): Corpus Val.Es.Co 3.0. Available online at <http://www.valesco.es>.

Méndez Guerrero, Beatriz (2015). Corpus Oral Juvenil del Español de Mallorca (COJEM). Available online at <https://ebuah.uah.es/dspace/handle/10017/25298>.

Notes

¹ Esta publicación es parte del proyecto de I+D+i «Estrategias pragmático-retóricas en la interacción conversacional conflictiva entre íntimos y conocidos: intensificación, atenuación y gestión interaccional (ESPRINT)» (ref. PID2020-114805GB-I00), financiado por MICIU/AEI/10.13039/501100011033/ y de la «Red Temática sobre comunicación conflictiva y mediación: interacción, vínculos relacionales y cohesión social» (CoCoMInt), financiada por la ayuda RED2022-134123-T de la convocatoria «Redes de Investigación 2022», MICIU/AEI /10.13039/501100011033. [Back]

² All translations of corpora examples are by the author. Retrospective imperatives and their translations are marked in bold; other phenomena of interest are underlined. [Back]

³ Most of the examples extracted in this research come from European Spanish, although corpora from various Spanish-speaking regions have been consulted. However, further investigation on this point is necessary. [Back]

⁴ Within the theory of construction grammar (Goldberg, 1965) a construction is a semi-compositional grammatical structure that can have various degrees of complexity and carries a certain interpretation. The form and the meaning of the construction is shaped by language usage. [Back]

⁵ Our translation. [Back]

⁶ In syntax, interdependency refers to a syntactic relationship between sentences situated between coordination and subordination. Refer to Gutiérrez Ordóñez (1997) for further details. [Back]

⁷ Personal translation. [Back]