1.Introduction

Being a cognitively oriented framework, relevance theory has not been extensively applied to the analysis of politeness phenomena. The few exceptions include the work of Jary (1998)Jary, Mark 1998 “Relevance Theory and the Communication of Politeness.” Journal of Pragmatics 30: 1–19. , Escandell-Vidal (1998Escandell-Vidal, Victoria 1998 “Politeness: A Relevant Issue for Relevance Theory.” Revista Alicantina de Estudios Ingleses 11: 45–57. , 2004Escandell-Vidal, Victoria 2004 “Norms and Principles: Putting Social and Cognitive Pragmatics Together.” In Current Trends in the Pragmatics of Spanish, ed. by Rosina Márquez-Reiter, and María Elena Placencia, 347–371. Amsterdam: John Benjamins. ), Padilla Cruz (2007)Padilla Cruz, Manuel 2007 “Politeness: Always Implicated?” In International Perspectives on Gender and Language, ed. by José Santaemilia, Patricia Bou, Sergio Maruenda, and Gora Zaragoza, 350–372. València: University of València. and Mazarella (2015)Mazzarella, Diana 2015 “Politeness, Relevance and Scalar Inferences.” Journal of Pragmatics 79: 93–106. . This paper makes a contribution to this largely unchartered territory by offering an account of addressative forms in Polish, focusing on the distinction between those that encode closeness between interlocutors (ty + a 2nd person verb form; an equivalent of the T form in the T/V opposition) and those that encode deference (pan/pani + a 3rd person singular verb; an equivalent of the V form in the T/V opposition). The account proposed herein relies solely on the already available relevance-theoretic toolkit and offers an extension of the treatment of pronouns to the realm of social deixis, as defined by Levinson (1979)Levinson, Stephen C. 1979 “Pragmatics and Social Deixis: Reclaiming the Notion of Conventional Implicature.” Proceedings of the Fifth Annual Meeting of the Berkeley Linguistics Society: 206–223. . By confining the scope of the paper to a single politeness-related phenomenon, I intend to align with the trend in 21st-century politeness research that prioritizes first-order politeness phenomena over second-order theoretical constructs (see Terkourafi 2019Terkourafi, Marina 2019 “Im/politeness: A 21st Century Appraisal.” Foreign Languages and Their Teaching 1 (6): 1–17. for an overview).

The basic notion employed in this analysis will be that of procedural meaning, introduced to relevance theory by Blakemore (1987)Blakemore, Diane 1987 Semantic Constraints on Relevance. Oxford: Blackwell. in her account of discourse markers, such as but, after all and others, and subsequently applied to a wide spectrum of communicative phenomena. Despite its broad usefulness, however, procedural meaning has remained notoriously difficult to grasp and still continues to be defined through contrast to other notions:11.A positive definition of procedural meaning was offered by Escandell-Vidal (2017)Escandell-Vidal, Victoria 2017 “Notes for a Restrictive Theory of Procedural Meaning.” In Doing Pragmatics Interculturally: Cognitive, Philosophical, and Sociopragmatic Perspectives, ed. by Rachel Giora, and Michael Haugh, 79–96. Berlin, Boston: De Gruyter/Mouton. , which however hinges on a more restrictive view of the phenomenon than that discussed below. See also Section 2. it is believed not to correspond to a conceptual representation, to be unparaphrasable, to lack equivalents in other languages, to be not accessible to introspection, etc. It was Wharton (2003Wharton, Tim 2003 “Interjections, Language and the ‘Showing–Saying’ Continuum.” Pragmatics and Cognition 11: 39–91. , 59) who delineated its scope by stating that procedural items activate “certain types of representations, or contextual assumptions, or expectations about cognitive effects”, and then Wilson added some precision to the definition of procedural expressions, postulating that they “are systematically linked to states of language users” (Wilson 2011Wilson, Deirdre 2011 “Procedural Meaning: Past, Present, Future.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 3–31. Bingley: Emerald Group Publishing. , 10), in contradistinction to conceptual expressions (e.g. dog, jump, happy) being “systematically linked to concepts, which are constituents of a language of thought”. In a later paper, Wilson (2016)Wilson, Deirdre 2016 “Reassessing the Conceptual–Procedural Distinction.” Lingua 175–176: 5–19. added that the function of procedural expressions is to activate domain-specific procedures which may be exploited in inferential communication. Wilson’s work on procedural meaning in relation to the model of the massively modular mind espoused by Sperber (1994Sperber, Dan 1994 “The Modularity of Thought and the Epidemiology of Representations.” In Mapping the Mind: Domain Specificity in Cognition and Culture, ed. by Lawrence Hirschfield, and Susan Gelman, 39–67. Cambridge: CUP. , 2001aSperber, Dan 2001a “In Defense of Massive Modularity.” In Language, Brain and Cognitive Development: Essays in Honor of Jacques Mehler, ed. by Emmanuel Dupoux, 47–57. Cambridge, MA: MIT Press.) and adopted in relevance theory is what provided much of the inspiration for the analysis in the present paper, in which it is assumed that the T and V pronouns encode procedures activating what Wilson (2016)Wilson, Deirdre 2016 “Reassessing the Conceptual–Procedural Distinction.” Lingua 175–176: 5–19. calls the “social cognition module”, and contribute to the hearer’s assessment of an utterance as socially appropriate or inappropriate. This may also lead to some expressive effects, such as distancing oneself from or belittling an interlocutor. In this way, on the basic level, the procedure activated by a T or V pronoun would be associated with a politic behaviour, whereas the extra layer of expressive meaning, if present, could be identified with politeness or impoliteness in the sense of Watts (2003)Watts, Richard 2003 Politeness. Cambridge: CUP. . Unless this distinction is specifically invoked, the term “politeness” will be used below in the general sense of linguistic politeness, which includes politic uses as well.

This paper is organized as follows. In the next section I characterize procedural meaning, focusing mainly on its features and also on where it is situated with respect to the massive modular model of the architecture of the human mind adopted in relevance theory. Then I go on to present the relevance-theoretic work on personal pronouns as procedural expressions. This is followed by a presentation of the system of Polish addressative forms and an analysis of the data, intended to support a procedural account of the deference form and the closeness form. In conclusion I address the question of whether the analysis proposed in this paper can be extended to other T/V languages.

2.Procedures and the modularity of mind

2.1Characteristic features of procedural items

The notion of procedural meaning, introduced to relevance theory by Blakemore (1987)Blakemore, Diane 1987 Semantic Constraints on Relevance. Oxford: Blackwell., was largely inspired by research on discourse connectives, such as but or so, which appeared to encode processing constraints, hence procedures, rather than elements of truth-conditional meaning. Initially, the scope of procedural meaning seemed to be restricted to the phenomena termed “conventional implicature” in Gricean pragmatics, but the following years saw its expansion into many other areas of communication, both linguistic and non-linguistic. In her paper on the heterogeneity of procedural meaning, Carston (2016)Carston, Robyn 2016 “The Heterogeneity of Procedural Meaning.” Lingua 175–176: 154–166. distinguishes four stages of its development. Within these stages, procedural accounts were offered for: (1) various extra-sentential devices marking logical relations between propositions, propositional attitude, speech act description and the like; (2) elements within propositions communicated, most notably pronouns with a referential function (Wilson and Sperber 1993Wilson, Deirdre, and Dan Sperber 1993 “Linguistic Form and Relevance.” Lingua 90: 1–25. ) and inflectional morphemes (Escandell-Vidal and Leonetti 2011Escandell-Vidal, Victoria, and Manuel Leonetti 2011 “The Rigidity of Procedural Meaning.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 81–102. Bingley: Emerald Group Publishing. ); (3) markers of emotive attitude, such as expletives, prosody and other conventionalized paralinguistic signals (Wharton 2003Wharton, Tim 2003 “Interjections, Language and the ‘Showing–Saying’ Continuum.” Pragmatics and Cognition 11: 39–91. ); (4) all conceptual items whose meaning needs to be adjusted to the context. Much as the first three are now accepted as standard within relevance theory, the fourth should be seen as a tentative proposal. But even the phenomena covered by stages 1–3 are diverse to the extent that it is impossible to delineate a common core for all of them. Several properties have been named, which may be shared by procedural items to a different degree. Following Carston (2016)Carston, Robyn 2016 “The Heterogeneity of Procedural Meaning.” Lingua 175–176: 154–166. and focusing on personal pronouns, which tend to be exceptional in some respects, these are the properties at issue: (1) introspective inaccessibility, which makes procedural items hardly paraphrasable. This is evidenced in rather complex analyses aimed to tease out the meaning of particular discourse connectives (e.g. Unger 2012aUnger, Christoph 2012a “Procedural Semantics, Metarepresentation, and Some Particles in Behdini Kurdish.” Lingua 122: 1613–1635. ; Casson 2020Casson, Sarah 2020 “The Greek Connective gar: Different Genres, Different Effects?” In Relevance Theory, Figuration, and Continuity in Pragmatics, ed. by Agnieszka Piskorska, 95–119. Amsterdam: John Benjamins. ; Lubberger 2020Lubberger, Beate 2020 “Metarepresentation Markers in Indus Kohistani: A Study with Special Reference to the Marker of Desirable Utterances loo .” In Relevance Theory, Figuration, and Continuity in Pragmatics, ed. by Agnieszka Piskorska, 121–164. Amsterdam: John Benjamins. ). Personal pronouns are indeed exceptional in this respect, being fairly easy to paraphrase or conceptualize (Carston 2016Carston, Robyn 2016 “The Heterogeneity of Procedural Meaning.” Lingua 175–176: 154–166. , 159); (2) non-compositionality, which means that procedural items do not enter easily into syntactic patterns with other items. Again, this does not apply rigorously to personal pronouns, which may enter into such patterns even in a creative way; (3) rigidity, originally discussed by Escandell-Vidal and Leonetti (2011)Escandell-Vidal, Victoria, and Manuel Leonetti 2011 “The Rigidity of Procedural Meaning.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 81–102. Bingley: Emerald Group Publishing. , by virtue of which procedural meaning (which does not undergo contextual modifications) will always impose its conditions when it conflicts with conceptual information or contextual inference (which are easily susceptible to modification). For instance, grammatical aspect always prevails over lexical aspect, as in I’m loving you, where the progressive aspects coerces the stative verb love into assuming a dynamic meaning, and the sense of definiteness imposed by the always forces the hearer to interpret an NP as definite, even in the absence of a definite referent in their cognitive environment, as when the sentence The dog was asleep would appear at the beginning of a short story, thereby creating the illusion that the reader is in the middle of the plot.

Carston (2016)Carston, Robyn 2016 “The Heterogeneity of Procedural Meaning.” Lingua 175–176: 154–166. mentions two more features, which can be seen as consequent upon rigidity: (4) not susceptible to non-literal use – unlike discourse connectives and the like, pronouns can to some extent be used non-literally, as their denotations can be broadened along the dimensions of gender or person (see the discussion of Scott 2016Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. below).22.Since such broadenings differ from the broadenings of concepts resulting in metaphors, I will follow Scott (2016)Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. in referring to them as ‘expressive uses’ rather than ‘non-literal uses’. Carston (2016)Carston, Robyn 2016 “The Heterogeneity of Procedural Meaning.” Lingua 175–176: 154–166. observes that personal pronouns are also susceptible to metarepresentational parodic use, a salient example of which would be the use of the plural we to ridicule someone’s aspirations of sounding royal. Interestingly, the parodic use seems not to be restricted to pronouns, since it is possible to imitate and exaggerate other facets of style relying on procedural devices, including intonation, facial expressions, and even one’s penchant for using specific discourse markers; (5) not polysemous – it is true that procedural items are not susceptible to developing new senses in the same way as words encoding concepts, even though discourse markers or mood indicators often display a number of related functions. Also, the V-type addressative forms analysed below remain in a kind of polysemy relation with nouns, from which they are derived (see Section 4 for details). That said, it can be concluded that polysemy among procedural items is indeed highly restricted.

Of all the features of procedural items discussed above, rigidity is probably the most stable, with the other characteristics merely tending to occur in them. As Carston (2016Carston, Robyn 2016 “The Heterogeneity of Procedural Meaning.” Lingua 175–176: 154–166. , 161) concedes, “it looks unlikely that there is any watertight test for telling whether some element of encoded meaning is conceptual or procedural”. This observation and the very heterogeneity of procedural items could undermine the theoretical status of the distinction between conceptual and procedural encoding and its empirical usefulness for analysis. A solution is offered by Escandell-Vidal (2017)Escandell-Vidal, Victoria 2017 “Notes for a Restrictive Theory of Procedural Meaning.” In Doing Pragmatics Interculturally: Cognitive, Philosophical, and Sociopragmatic Perspectives, ed. by Rachel Giora, and Michael Haugh, 79–96. Berlin, Boston: De Gruyter/Mouton. , who maintains that, first of all, the domain of procedural meaning should be restricted to linguistic items with the exclusion of non-linguistic signs. Then, the distinction can be neatly drawn along the divide between lexical and functional items, with the former corresponding to conceptual and the latter to procedural meaning (with the further caveat that not all functional items necessarily contribute procedural meaning). She also claims that delineating the class in question so that it includes only linguistic items is conducive to providing a fairly specific definition of procedural meaning in terms of three criteria: being an attribute of a functional category, ability to be modelled as computational algorithm, and ability to operate on the conceptual – intentional systems (Escandell-Vidall 2017Escandell-Vidal, Victoria 2017 “Notes for a Restrictive Theory of Procedural Meaning.” In Doing Pragmatics Interculturally: Cognitive, Philosophical, and Sociopragmatic Perspectives, ed. by Rachel Giora, and Michael Haugh, 79–96. Berlin, Boston: De Gruyter/Mouton. , 84).

Escandell-Vidal’s (2017)Escandell-Vidal, Victoria 2017 “Notes for a Restrictive Theory of Procedural Meaning.” In Doing Pragmatics Interculturally: Cognitive, Philosophical, and Sociopragmatic Perspectives, ed. by Rachel Giora, and Michael Haugh, 79–96. Berlin, Boston: De Gruyter/Mouton. restrictive view of procedural meaning does indeed offer important theoretical benefits for the construal of procedural meaning in relevance theory. It should be noted that the material analysed in this paper, i.e. the Polish T and V addressative forms, is compatible with the restrictive view, as the items in question are linguistic. On the other hand, as will be described in detail below, the procedures associated with the T/V-politeness element may not target the inferential comprehension system. In the next section we will consider how procedurality can be grafted on the modular model of the human mind, in which comprehension is seen as distinct from argumentation, believability or social dimensions of communication.

2.2Massive modularity and procedural items

In the original formulation of the framework (Sperber and Wilson 1986Sperber, Dan, and Deirdre Wilson 1986 Relevance: Communication and Cognition. Oxford: Blackwell.), the authors follow Fodor’s (1983)Fodor, Jerry A. 1983 The Modularity of Mind. Cambridge, MA: MIT Press. view on the architecture of the human mind, which is seen as consisting of a central processor responsible for carrying out higher cognitive functions and a number of modular domain-specific input systems. Later, this view becomes abandoned in favour of the “massive modularity” model (Sperber 1994Sperber, Dan 1994 “The Modularity of Thought and the Epidemiology of Representations.” In Mapping the Mind: Domain Specificity in Cognition and Culture, ed. by Lawrence Hirschfield, and Susan Gelman, 39–67. Cambridge: CUP. , 2001aSperber, Dan 2001a “In Defense of Massive Modularity.” In Language, Brain and Cognitive Development: Essays in Honor of Jacques Mehler, ed. by Emmanuel Dupoux, 47–57. Cambridge, MA: MIT Press.), in which the mind is seen as consisting only of informationally encapsulated modules specialized in performing tasks specific to their respective domains. There are certainly questions about massive modularity that have yet to be addressed,33.One of the most troubling issues consequent upon adopting the massively modular model of the mind is that of the self or consciousness – see Mercier and Sperber (2009)Mercier, Hugo, and Dan Sperber 2009 “Intuitive and Reflective Inferences.” In In Two Minds: Dual Processes and Beyond, ed. by Jonathan S. B. T. Evans, and Keith Frankish, 149–170. Oxford: OUP. . Another, which is more closely related to pragmatics, is to what extent and how the independent and encapsulated modules collaborate to yield optimally relevant interpretations (Piskorska 2016Piskorska, Agnieszka 2016 “Perlocutionary Effects and Relevance Theory.” In Relevance Theory: Recent Developments, Current Challenges and Future Directions, ed. by Manuel Padilla Cruz, 287–305. Amsterdam: John Benjamins. ). but since this view has now become standard in relevance theory, some discussion is due on the consequences of its adoption for the account of procedural meaning. As mentioned above, Blakemore (1987)Blakemore, Diane 1987 Semantic Constraints on Relevance. Oxford: Blackwell. approaches discourse connectives as procedural constraints on the comprehension process. An increasing focus on argumentation and cognitive mechanisms aimed at protecting an individual against deception (Sperber et al. 2010Sperber, Dan, Fabrice Clément, Christoph Heintz, Olivier Mascaro, Hugo Mercier, Gloria Origgi, and Deirdre Wilson 2010 “Epistemic Vigilance.” Mind and Language 25: 359–393. ) has led to rethinking the role of connectives in relevance theory, which are seen as devices affecting the believability of an assumption rather than its comprehension (Sperber 2001bSperber, Dan 2001b “An Evolutionary Perspective on Testimony and Argumentation.” Philosophical Topics 29: 401–413. ). Remarking on this idea, Wilson (2011Wilson, Deirdre 2011 “Procedural Meaning: Past, Present, Future.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 3–31. Bingley: Emerald Group Publishing. , 37) observes that “it clearly opens up the possibility of an alternative to the standard relevance-theoretic account, on which the procedures encoded by discourse connectives have less to do with understanding than with believing”.

Wilson (2016)Wilson, Deirdre 2016 “Reassessing the Conceptual–Procedural Distinction.” Lingua 175–176: 5–19. elaborates on this idea, adding that discourse connectives may in fact play a role in both comprehension and argumentation. She provides further examples of procedural devices that typically activate various modules related to, but existing independently of communication. These include: the emotion-reading module, which is sensitive to affective intonation, interjections and attitudinal particles (Wharton 2003Wharton, Tim 2003 “Interjections, Language and the ‘Showing–Saying’ Continuum.” Pragmatics and Cognition 11: 39–91. , 2009Wharton, Tim 2009 Pragmatics and Non-Verbal Communication. Cambridge: Cambridge University Press. ); the epistemic vigilance module, sensitive to expressions of epistemic modality and evidentiality (Wilson 2011Wilson, Deirdre 2011 “Procedural Meaning: Past, Present, Future.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 3–31. Bingley: Emerald Group Publishing. ; Unger 2012bUnger, Christoph 2012b “Epistemic Vigilance and the Function of Procedural Indicators in Communication and Comprehension.” In Relevance Theory: More than Understanding, ed. by Ewa Wałaszewska, and Agnieszka Piskorska, 45–73. New Castle upon Tyne: Cambridge Scholars Publishing.); or the social cognition module, which should be naturally linked to facework and politeness. The idea that procedural devices may activate various mechanisms, not necessarily directly involving comprehension, has important consequences for defining the kind of effects they bring about. Much as procedures targeted at understanding a message (i.e. the comprehension module) yield various components of the utterance’s explicit and implicit import, other procedures result in effects of a different kind, such as attribution of emotions, mental states or allocation of trust (Wilson 2016Wilson, Deirdre 2016 “Reassessing the Conceptual–Procedural Distinction.” Lingua 175–176: 5–19. ). When it comes to procedures targeting the social cognition module, it can be assumed that they lead to the recognition of the speaker’s social competence. This assumption squares well with Fraser’s (1990)Fraser, Bruce 1990 “Perspectives on Politeness.” Journal of Pragmatics 14: 219–236. and Jary’s (1998)Jary, Mark 1998 “Relevance Theory and the Communication of Politeness.” Journal of Pragmatics 30: 1–19. objection to Brown and Levinson’s (1987)Brown, Penelope, and Stephen C. Levinson 1987 Politeness: Some Universals in Language Usage. Cambridge: CUP. model of politeness, in which the speaker’s being polite should surface as an implicature, and with Jary’s (1998)Jary, Mark 1998 “Relevance Theory and the Communication of Politeness.” Journal of Pragmatics 30: 1–19. proposal that politeness resulting from the use of unmarked polite forms should be seen not as communicated, but rather as anticipated. By this token, it is assumed here that the hearer’s recognition of a certain linguistic form as adequate from a social point of view is not a component of an utterance’s meaning, neither on the explicit nor on the implicit level.

An independent strand of relevance-theoretic research supporting the above-stated point was contributed by Escandell-Vidal (2004)Escandell-Vidal, Victoria 2004 “Norms and Principles: Putting Social and Cognitive Pragmatics Together.” In Current Trends in the Pragmatics of Spanish, ed. by Rosina Márquez-Reiter, and María Elena Placencia, 347–371. Amsterdam: John Benjamins. , who discusses the difference between inference-based pragmatics, explaining various levels of communicated meaning, and norm-based pragmatics, focusing on social effects of communication. Since it is unfeasible on theoretical and empirical grounds to subsume both under one category, as the former is governed by cognitive efficiency and the latter stems from generalizations drawn over social preferences observed in certain groups, Escandell-Vidal (2004)Escandell-Vidal, Victoria 2004 “Norms and Principles: Putting Social and Cognitive Pragmatics Together.” In Current Trends in the Pragmatics of Spanish, ed. by Rosina Márquez-Reiter, and María Elena Placencia, 347–371. Amsterdam: John Benjamins. postulates a division of labour between an inference-driven comprehension module and a norm-driven social categorization system, with this essentially modular approach to pragmatics being the only viable way of integrating its social and cognitive branches. Although the main goal of Escandell-Vidal’s chapter is to sketch an integrated architecture of pragmatic theory, rather than the modular architecture of the mind, her contribution should be recognized as providing a solid background for later work developing the idea that various linguistically encoded elements can in fact target different modules. If procedural expressions are, as Wilson (2011)Wilson, Deirdre 2011 “Procedural Meaning: Past, Present, Future.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 3–31. Bingley: Emerald Group Publishing. puts it, systematically linked to states of language users, then Escandell-Vidal’s (2004)Escandell-Vidal, Victoria 2004 “Norms and Principles: Putting Social and Cognitive Pragmatics Together.” In Current Trends in the Pragmatics of Spanish, ed. by Rosina Márquez-Reiter, and María Elena Placencia, 347–371. Amsterdam: John Benjamins. social categorization system, governed by norms acquired over one’s life in social interactions, should be equated with the social module postulated later by Wilson (2016)Wilson, Deirdre 2016 “Reassessing the Conceptual–Procedural Distinction.” Lingua 175–176: 5–19. . This, together with a procedural account of personal pronouns, provides the background against which the T/V addressative forms can be analyzed. First, however, it is necessary to recapitulate some details of the relevance-theoretic stance on personal pronouns.

3.Pronouns as procedural and truth-functional items

Wilson and Sperber (1993)Wilson, Deirdre, and Dan Sperber 1993 “Linguistic Form and Relevance.” Lingua 90: 1–25. propose to classify pronouns as procedural and at the same time truth-conditional. In so doing, they draw on observations by Kaplan (1989)Kaplan, David 1989 “Demonstratives.” In Themes from Kaplan, ed. by. Joseph Almog, John Perry, and Howard Wettstein, 481–563. Oxford: OUP., who claims that analysing the pronoun I as meaning ‘the speaker’ vs. as carrying an instruction to identify the speaker makes a difference to the truth-conditions of an utterance. On this account, “a pronoun might activate a certain class of candidate referents from which the hearer must choose” (Wharton 2003Wharton, Tim 2003 “Interjections, Language and the ‘Showing–Saying’ Continuum.” Pragmatics and Cognition 11: 39–91. , 59).

Scott (2016)Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. points out that one of the specific problems that needs to be addressed in a procedural account of pronouns is how the procedural information related to picking out a referent interacts with such grammatical categories as gender, person and number marked on a pronoun, which could be seen as including conceptual information. Some ideas of how the combination of conceptual and procedural elements can be handled are offered by de Saussure (2011)de Saussure, Louis 2011 “On Some Methodological Issues in the Conceptual/Procedural Distinction.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 55–79. Bingley: Emerald Group Publishing. , who claims that conceptual information is a parameter on which the procedure operates, which makes the former part of, and dependent on, the latter. In a similar vein, Curcó (2011)Curcó, Carmen 2011 “On the Status of Procedural Meaning in Natural Language.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 33–54. Bingley: Emerald Group Publishing. postulates that procedures specify how conceptual representations are to be processed and are themselves sealed off, or “bracketed”, from the level of consciousness and conceptual representations. Scott, on the other hand, discards the view that pronouns are a blend of procedural and conceptual information, arguing for a fully procedural account. Even if the ideas of number or gender as such lend themselves to conceptual analysis, Scott (2016)Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. argues, it is unlikely that they figure in the explicature of an utterance containing the pronoun they (even if this pronoun has a plural referent) or the pronoun she (even if it has a female referent). Indeed, if the encoded procedural meaning is represented at a sub-personal level, as Scott proposes, it does not enter the basic-level or higher-level explicatures of an utterance at all. Instead, it should be seen as a “pointer” to a procedure activating a search for referents that meet the conditions specified by the encoding element. In fact, even though de Saussure’s (2011)de Saussure, Louis 2011 “On Some Methodological Issues in the Conceptual/Procedural Distinction.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 55–79. Bingley: Emerald Group Publishing. and Curcó’s (2011)Curcó, Carmen 2011 “On the Status of Procedural Meaning in Natural Language.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 33–54. Bingley: Emerald Group Publishing. views, later endorsed by Escandell-Vidal (2017)Escandell-Vidal, Victoria 2017 “Notes for a Restrictive Theory of Procedural Meaning.” In Doing Pragmatics Interculturally: Cognitive, Philosophical, and Sociopragmatic Perspectives, ed. by Rachel Giora, and Michael Haugh, 79–96. Berlin, Boston: De Gruyter/Mouton. , appear to differ from those espoused by Scott (2016)Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. in a number of details, they all boil down to the observation that the elements of conceptual representations to which procedures are sensitive, such as gender, are not part of the explicit or implicit import of an utterance. They remain within the scope of procedures, which, according to Wilson (2011Wilson, Deirdre 2011 “Procedural Meaning: Past, Present, Future.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 3–31. Bingley: Emerald Group Publishing. , 11), “are generally seen as formulated in a sub-personal ‘machine language’ distinct from the language of thought”.

Pronouns may play other roles than reference assignment. Scott (2016)Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. scrutinizes their potential to convey expressive meaning, which, although limited, may be quite powerful. For instance, referring to someone else’s child as it may have a belittling effect; referring to a car as she may expresses affect. Clearly, this feature of pronouns makes them stand out from other procedural items, as mentioned in the previous section. It seems that when T/V pronouns appear in their addressative function, their expressive potential is enhanced, as apart from the gender dimension, the social deixis parameter of closeness/deference can also be exploited for that purpose. This will be elaborated on in Section 4.

4.Addressative forms in Polish as expressions of social deixis

Pronouns have been traditionally considered as deictic items, which means that their interpretation in an utterance hinges on the knowledge of some aspects of the communicative act in which this utterance occurs (Fillmore 1997Fillmore, Charles. J. 1997 Santa Cruz Lectures on Deixis. Stanford: CSLI Publications.). The extant procedural treatment of personal pronouns developed within relevance theory thus far has focused on person deixis. The present paper, in contrast, focuses on social deixis, defined by Levinson (1979Levinson, Stephen C. 1979 “Pragmatics and Social Deixis: Reclaiming the Notion of Conventional Implicature.” Proceedings of the Fifth Annual Meeting of the Berkeley Linguistics Society: 206–223. , 206) as “those aspects of language structure that are anchored to the social identities of participants, or to relations between them, or to relations between them and other referents” including “honorifics, titles of address, second person pronominal alternates and associated verb agreements”. Levinson distinguishes two types of social dimensions encoded by deictic expressions, namely absolute and relational. The absolute dimension pertains to forms reserved for specific individuals, such as Mr. President. The relational dimension may refer to relations between various participants in a speech event, such as the speaker and a third-party referent or the speaker and the addressee. In the interest of clarity, I will limit the discussion to the relation between the speaker and the addressee.

Let us now turn to the characteristics of the Polish addressative forms, which do not align neatly with the set of personal pronouns. Since Polish is a pro-drop language, the second-person singular pronoun ty, marked for an informal or close type of relationship between interlocutors, does not typically surface as an independent syntactic unit, unless used emphatically. In the absence of ty, the grammatical category of second person singular is encoded by the inflectional ending of a verb. Both the pronominal form ty and the corresponding verb ending are non-ambiguously used for marking the second person singular. In the plural, the situation is more complicated, as there is no pronominal form corresponding to the V category. Instead, this category covers a class of nouns that can function as a deference pronoun, among which the most commonly used are pan/pani (‘sir’/‘madam’), with others denoting titles, such as ‘doctor’, ‘professor’, ‘vicar’, etc., combining with a third-person singular verb form. The use of various Polish addressative forms is illustrated below, with English gloss and free translations (the gloss translation is given only for the items relevant for the current analysis; the forms pan/pani are represented as such in the glosses).

As mentioned above, ty may surface when the speaker wishes to highlight the addressee for emphasis or other discourse-related reasons:

The deference form is illustrated below:

The formal nuances related to the specifics of Polish grammar should not impact on the main line of argumentation presented in this paper, which is premised on the assumption taken from Łaziński (2006Łaziński, Marek 2006 O panach i paniach: Polskie rzeczowniki tytularne i ich asymetria rodzajowo–płciowa (On Ladies and Gentlemen: Polish Titulary Nouns and their Gender Asymmetry). Warsaw: PWN., 15) that in functional terms, pan/pani as addressative forms can be approached as pronouns corresponding to the V form. In principle, the nominal and conceptual reading of pan/pani, corresponding to English ‘lady’/‘gentleman’, should not pose a problem for this analysis in the light of the research discussed in Section 3, where solutions were presented to potential challenges posed by the combination of conceptual and procedural encoding within one linguistic item. However, since the presence of the conceptual element in pronominal pan/pani may be felt to be stronger than in he/she due to their nominal origin, an additional justification should be offered for the claim that the conceptual element does not surface in an utterance’s explicit or implicit import. For this purpose, let us resort to analogy with other items which have diachronically related polysemes, and for which compelling procedural analyses have been offered, such as still (Higashimori 1992Higashimori, Isao 1992 “BUT/YET/STILL and Relevance Theory.” In Papers Presented to Professor Yoshimitsu Narita on the Occasion of his Sixtieth Birthday, 333–354. Tokyo: Eihosha.) or well (Jucker 1993Jucker, Andreas H. 1993 “The Discourse Marker Well: A Relevance-Theoretical Account.” Journal of Pragmatics 19: 435–452. ). In his analysis of grammaticalization as a shift from conceptual to procedural encoding, Nicolle (1998)Nicolle, Stephen 1998 “A Relevance Theory Perspective on Grammaticalization.” Cognitive Linguistics 9: 1–35. addresses the phenomenon of “semantic retention” or “residual lexical meaning” in procedural items, as in the use of conceptual well in all is well, motivating the function of procedural well (which could be approximated as ‘all that has been said so far has been well received and noted’). A similar point on grammaticalization was made by de Saussure (2011)de Saussure, Louis 2011 “On Some Methodological Issues in the Conceptual/Procedural Distinction.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 55–79. Bingley: Emerald Group Publishing. on the basis of his analysis of French conjunctions such as puisque, parce que, and others, the conceptual meaning of which has become opaque over time. Thus, the polysemy of pan/pani, with one of their readings being conceptual and the other procedural, is not an isolated case and as such should not raise controversy. Even if the presence of ‘ladyness’ and ‘gentlemanhood’ as conceptual elements can be felt in the addressative use of pan/pani, it is a residual kind of lexical meaning, whose role is confined to providing a cognitive rationale for the use of these items, which are marked for gender and deference. This residual meaning plays a role in explaining the affinity of procedural items with their etymologically related conceptual counterparts, but does not enter into comprehension processes and is not represented in the explicature of an utterance.

The claim that the gender of the addressative form is not represented conceptually in the explicature of an utterance can be further strengthened by observations on grammatical gender in those languages that mark it obligatorily on nominals. In Polish, for instance, nouns marked for the feminine gender (the ending -a) include szafka (‘cupboard’), filiżanka (‘cup’), herbata (‘tea’), and it would be rather implausible to insist that a biological element of femininity is represented in these concepts. Needless to say, the situation is different with animate nouns and natural gender, but even if the value of this observation is limited, it does show that gender as a category does not have to surface in conceptualizations of nominals.

Having argued that the conceptual element in pan/pani does not contribute to the import of an utterance, even despite the connection of pan/pani to nouns with conceptual content, I now turn to details of the procedural encoding of the politeness element. Brown and Gilman (1960)Brown, Roger, and Albert Gilman 1960 “The Pronouns of Power and Solidarity.” In Style in Language, ed. by Thomas Sebeok, 253–276. London and New York: The Technology Press of Massachusetts Institute of Technology and John Wiley & Sons, Inc. explain the semantics of the T/V distinction in terms of two parameters: power and solidarity, with the power parameter being more prominent in class-differentiated societies, and the solidarity parameter determining the use of T pronouns. The use of the T addressative form is then said to express solidarity between the speaker and the addressee, whereas the use of the V form lacks this feature. Due to the fact that the term ‘solidarity’ seems to have strong associations with mutual support and group identity, a better candidate for capturing said kind of relation between the speaker and addressee seems to be ‘closeness’. In terms of procedural meaning, it can thus be postulated that the procedure encoded by ty is ‘mark the addressee as close to the speaker’, and the procedure encoded by pan/pani would logically follow as ‘mark the addressee as not close to the speaker’. The addressee’s social cognition module, which these procedures target, assesses whether a given form has been used appropriately on the basis of generalizations and observations gathered in the course of their interactions with other members of society. In accordance with previously made claims, if the use of an addressative form is deemed felicitous by the social cognition module, nothing is added to the import of an utterance, which results in a politic use (Watts 2003Watts, Richard 2003 Politeness. Cambridge: CUP. ) or satisfying anticipated politeness (Jary 1998Jary, Mark 1998 “Relevance Theory and the Communication of Politeness.” Journal of Pragmatics 30: 1–19. ). In terms of Brown and Levinson’s (1987)Brown, Penelope, and Stephen C. Levinson 1987 Politeness: Some Universals in Language Usage. Cambridge: CUP. politeness theory, such a use of the pronoun does not pose a threat to the hearer’s face.

Misuses of addressative forms may occur due to the fact that the conventions related to social relations between interlocutors have recently diversified in Polish culture, and it is not always clear which norm should be followed. Much as the T/pan/pani distinction is standard and it would be rude to address an adult stranger with a T form in most situations, in some more progressive circles it is the T form that may be used straight away, for instance, when a new employee in a corporation is being introduced to their colleagues. If a speaker misjudges which norm is expected of them, they risk being considered rude, or conversely, old-fashioned. Unintentional non-compliance with the set of norms governing the use of the T/V forms can be equated with the speaker’s accountability for the conventionally established social meaning (cf. Haugh 2013Haugh, Michael 2013 “Speaker Meaning and Accountability in Interaction.” Journal of Pragmatics 48: 41–56. ), which explains why such acts may be received as face-threatening, even despite the lack of the speaker’s intention to threaten the hearer’s face.

Let us now turn to the features of procedural items enumerated in Section 2 in order to verify if and how they apply to the politeness-related procedures encoded by ty and pan/pani. Three of them have already been discussed in this section: as regards non-accessibility, it is observed that it is not displayed by pan/pani, for which the residual conceptual meaning is felt to be relatively strong; the polysemy of pan/pani also stems from their etymological link to the corresponding nouns; the property of non-compositionality is absent, with pan/pani being often combined with professional titles, and ty being connectable with any bare noun phrase. With respect to these three properties, the addressative forms are analogous to the non-T/V pronouns.

In what follows I will focus on the other two properties, i.e. rigidity and the potential of the T/V forms to serve expressive functions. The examples presented below involve intentional extensions of standard uses, which will provide an opportunity to discuss extra stylistic effects belonging to the domain of communicated meaning, rather than to the realm of ‘anticipated politeness’. As has been mentioned, the criterion of rigidity is met if the reading imposed by the procedure prevails over contextual information. What might be expected here is that the procedural meaning should win when combined with incompatible conceptual representations. This will be illustrated by two examples drawn from the discourse of fiction. One is the title of a popular comedy show aired on Polish radio for many years:

The juxtaposition of the confession ‘I love you’ with the deference form of address pan is incongruous (at least at any moment in time later than the 19th century). The incongruity would not arise if the procedure encoding social distance could be somehow adjusted to the conceptual meaning of love, denoting a close relationship. It is not, however, and the clash between intimacy and ‘non-closeness’ or deference lingers on. This is not fully parallel with the cases of rigidity discussed by Escandell-Vidal and Leonetti (2011)Escandell-Vidal, Victoria, and Manuel Leonetti 2011 “The Rigidity of Procedural Meaning.” In Procedural Meaning: Problems and Perspectives, ed. by Victoria Escandell-Vidal, Manuel Leonetti, and Aoife Ahern, 81–102. Bingley: Emerald Group Publishing. , where the procedural element, such as grammatical aspect, coerces the conceptual element, such as the lexical aspect, to modify its meaning. Here, no coercion occurs, possibly because of the fact that the procedure does not target the inferential comprehension system, but the procedural meaning can still be deemed rigid as it is not itself affected by contextual factors. The incongruity gives rise to the impression that the speaker (i.e. the fictional character in the show) is super-polite and timid, and this is the extra effect obtained by an intentional extension of the use of pan to a situation in which intimacy would be expected. Since this is an ostensive act, the extraordinary politeness or shyness may be represented in the higher-level explicature of the utterance as a communicative effect intended by the speaker, and its source can be identified with the clash between the ‘non-closeness’ procedure and other elements of the utterance.

The other illustrative example indicating rigidity comes from the movie Demony Wojny (‘Demons of War’), in which a major, who had ignored his superior’s orders, addresses an attorney in the following way:

Again, in this case the procedure encoding ‘non-closeness’ in the addressative form pan is not in any way adjusted or weakened by the presence of the strong profanity, which normally signals lack of distance between interlocutors. On the contrary, it is the contrast between the addressative and the f-word that is responsible for the impression that the ruthless major is extremely offensive to the attorney, despite the social distance and military hierarchy, which he, in a way, obeys. Like in the previous example, the speaker’s intention to offend the addressee may be represented in the higher-level explicature of the utterance as part of the communicated import of the utterance. Examples of conflict between procedurally encoded deference and conceptually encoded closeness are hard to come by. But it seems that (4) and (5) provide sufficient evidence for the fact that the procedure encoding non-closeness in pan/pani is rigid. For one thing, it does not undergo contextual modifications and when intentionally juxtaposed with non-matching conceptual content, it leads to incongruity, which in turn may result in communicative effects.

The use of T/V pronouns for expressive purposes, such as showing affection to an inanimate object or belittling a person, deserves special merit. The addressative forms marked for politeness seem to be even better candidates for expressing such effects than non-T/V pronouns, as discussed by Scott (2016)Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. . A straightforward case of this kind stems from the speaker using the ty form to a complete stranger, which typically expresses lack of respect and an intention to belittle the hearer. As is often the case, drivers and pedestrians can experience strong negative emotions concerning the behaviour of other people using vehicles or walking through the streets. On such occasions, the following rhetorical questions can be heard:

Unlike in (5), in (6) and (7) the lack of deference procedurally encoded by the dative form ci and the 2nd person singular ending, respectively, works hand in hand with the offensive character of those utterances, questioning one’s ability to drive or to obey the safety rules in the street. The violation of the social norm results in undermining the social status of the addressee, who is thereby addressed as someone not deserving a standard treatment.

Using the deference addressative form pan/pani can also have a negative effect on the hearer. Consider a situation in which two people, a man and a woman in their forties, who work in the same international company but are not closely acquainted with each other, meet incidentally in a coffee shop. One of them takes a seat near the other because the place is quite crowded. They exchange a few casual remarks on new fanciful coffee flavours and find out that both of them are drinking pumpkin flavoured coffee. They inquire about each other’s impression about the beverage:

As was mentioned before, different social norms exist simultaneously among various groups of people and it is not always clear to which norm an individual subscribes. Being employees of an international company, the interlocutors are likely to identify with the progressive norm, according to which people are expected to use the T form (expressed as the dative ci in [8a]) in their encounters with other members of the staff, without deploying the traditional ritual of switching from the V form to the T form only after having gotten sufficiently acquainted. On the other hand, among their age group the norm obliging speakers to use the V form by default may still hold strong. The man subscribes to the more progressive norm, whereas the woman clearly does not, which she signals with the use of pan in her utterance. This has a distancing and patronizing effect, potentially evoking a negative emotional response in the man, whose social competence is being tacitly questioned by an act of imposing the more conservative form on him.

Another situation in which using the V form has a putting-down effect has been attested in an interview, posted on an internet gossip site. A very young male interviewer talks to a female celebrity participating in a TV dance contest; the woman may be about twenty years his senior. The woman expresses her wish to be addressed with the ty form, but the interviewer persists in using the pani form. In this way, he creates an effect of excluding the interviewee from the group of the other contestants, who are typically addressed with the T form. Considering the fact that some present-day media subscribe to the cult of youth and many celebrities want to be considered young irrespective of their age, the use of the deference form additionally emphasized the age gap between the interlocutors, thereby adding ageist overtones to the interviewer’s utterances addressed to the celebrity.

It is not certain if in the above-described situation the interviewer uses the V form deliberately to create the effect of exclusion. In some types of discourse, however, the addressative forms are intentionally exploited to create a certain image of the interlocutor that suits the speaker’s current needs, with political discourse being a notorious case in point. Since the commonly accepted practice in some professional or institutional circles is to combine the form pan/pani with a title, the omission of the title has a belittling effect. Łaziński (2006)Łaziński, Marek 2006 O panach i paniach: Polskie rzeczowniki tytularne i ich asymetria rodzajowo–płciowa (On Ladies and Gentlemen: Polish Titulary Nouns and their Gender Asymmetry). Warsaw: PWN. labels this strategy as title deprivation, and as Kostro and Wróblewska-Pawlak (2013)Kostro, Monika, and Krystyna Wróblewska-Pawlak 2013 “Formy adresatywne jako środek jawnej i ukrytej deprecjacji kobiet polityków w polskim dyskursie polityczno-medialnym.” (“Addressative Forms as a Means of Overt and Covert Discrimination of Female Politicians in Polish Political and Media Discourse”). Tekst i Dyskurs 6: 153–168. argue, it is one of the strategies typically applied by some male politicians to undermine perceptions about competences of female politicians. When the title is omitted, the professional qualifications of a female politician are downplayed, thereby reducing her status to that of a woman per se, rather than a woman-MP, woman-commissioner, woman-chair, etc. In this case and in the previous ones (starting with Example [4]), the non-standard expressive uses are intentionally and ostensively exploited by the speaker, and as such are likely to become part of the communicated import of the utterance. Examples (5), (6), (7) and (8) will also be perceived as face-threatening acts not only by virtue of departing from the established convention, but also by virtue of the speaker’s intention.

This section has offered a procedural account of the Polish addressative forms ty/pan/pani grounded in previous relevance-theoretic research on pronouns. I have argued that the politeness-related element is encoded by a specific procedure targeting the social-cognition module. This is a fairly natural continuation of earlier attempts to link procedural meaning to politeness, such as Watts (2003)Watts, Richard 2003 Politeness. Cambridge: CUP. ; additionally, it combines ideas related to the modularity of mind, the division of labour between modules responsible for various aspects of pragmatic meaning (Escandell-Vidal 2004Escandell-Vidal, Victoria 2004 “Norms and Principles: Putting Social and Cognitive Pragmatics Together.” In Current Trends in the Pragmatics of Spanish, ed. by Rosina Márquez-Reiter, and María Elena Placencia, 347–371. Amsterdam: John Benjamins. ), and the role of procedures in meaning making (Wharton 2003Wharton, Tim 2003 “Interjections, Language and the ‘Showing–Saying’ Continuum.” Pragmatics and Cognition 11: 39–91. ; Wilson 2016Wilson, Deirdre 2016 “Reassessing the Conceptual–Procedural Distinction.” Lingua 175–176: 5–19. ; Escandell-Vidal 2017Escandell-Vidal, Victoria 2017 “Notes for a Restrictive Theory of Procedural Meaning.” In Doing Pragmatics Interculturally: Cognitive, Philosophical, and Sociopragmatic Perspectives, ed. by Rachel Giora, and Michael Haugh, 79–96. Berlin, Boston: De Gruyter/Mouton. ). I also devoted some space to the potential of the T/V distinction to express non-cognitive effects (or not purely cognitive effects), which draws on Scott’s (2016)Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. work on the expressive potential of personal pronouns, and which should also be seen against the background of other procedural items which have been described as specifically suited for such purposes (see Padilla Cruz 2020Padilla Cruz, Manuel 2020 “Towards a Relevance-Theoretic Approach to the Diminutive Morpheme.” Russian Journal of Linguistics 24: 774–795. on diminutive morphemes). When the use of an addressative form merely complies with social norms, it does not make a contribution to the communicated import of an utterance, whereas when exploited ostensively for the sake of obtaining interpersonal or stylistic effects, its contribution surfaces in the higher-level explicature and/or implicatures of an utterance.

5.Conclusion

The relevance-theoretic account of pronouns offered by Wilson and Sperber (1993)Wilson, Deirdre, and Dan Sperber 1993 “Linguistic Form and Relevance.” Lingua 90: 1–25. and Scott (2016)Scott, Kate 2016 “Pronouns and Procedures: Reference and Beyond.” Lingua 175–176: 69–82. is fairly exhaustive for the realm of person deixis, but a gap still exists concerning the T/V distinction as pertinent to social deixis and present in, among many other languages, Spanish, French, German and Slavic ones. This paper attempts to fill this gap by discussing the Polish addressative forms corresponding to the T/V distinction in terms of procedural meaning. The procedures associated with the respective pronominal forms are postulated to target the social-cognition module. The general character of the main claim advanced here – that the Polish T/V pronouns feature procedurally encoded politeness elements – should make it universally applicable to other languages which differentiate between the deference form and the closeness form. Needless to say, linguistic and cultural differences are only to be expected and can be addressed in separate research.