This article sets out to provide a summarized historical survey of some relevant contributions to axiological semantics from linguists included in the two most influential linguistic traditions from the 50s to the late 70s: structural linguistics and transformational-generative grammar. In fact, even if their representative opinions at that time could now be regarded as incidental, it would be unfair to claim that some of their intuitions and principles have not been crucial in illuminating essential aspects of the parameter of axiological evaluation in modern linguistics. In this respect, it is worth highlighting the ideas disseminated by key figures of the Saussurean or Chomskyan traditions from the perspective given by half a century of evolution in the field. For that purpose contributions are included from authors such as Apresjan, Bally, Coseriu, Ducháček, Fodor, Grzegorek, Guiraud, Katz, Klima, Pottier, Stati, and Ullmann, among others.
The focus in this chapter is on the evaluative systems of attitude, as modelled in the appraisal framework. While fully accepting the validity of the model, I explore three practical issues that arise in applying it. The first concerns the question of whether affect, the core area of attitude, should be distinguished from the more general field of emotion talk, and, if so, how. The second issue arises because two related but distinct criteria are used to distinguish judgements and appreciations: the kind of entity evaluated (the target) and the value ascribed to it. As a result, there may be an apparent mismatch between these two criteria. The third difficulty can be called the ‘Russian doll’ dilemma. An expression of one category of attitude may function as a token (an indirect expression) of a different category; and that token may itself function as an indirect expression of yet another category, and so on. This poses problems for a study of evaluation which relies on quantitative analysis. I suggest my own principles to help make life a little easier for anyone engaged in analysing evaluation in text.
Appraisal is a term referring to systems within Systemic Functional Linguistics that map evaluative language. This chapter reviews two Appraisal systems (Attitude and Graduation) and assesses their viability for analysing evaluation in narratives and students’ written responses to these. Within this context, a crucial affordance of Appraisal is its semantic orientation, which allows analysts to capture the delicate interweaving of explicit and implicit attitudes that powerfully positions readers of narratives. However, the sprawling, cumulative and context-sensitive nature of evaluation requires a rigorous methodology for using Appraisal for coding purposes. We address this challenge, proposing a cline of implicitness that relates degrees of reliability to instances of evaluation depending on their textual and/or cultural/institutional environments.
In this chapter we present the results of both a theoretical and empirical study on the evaluative character of verbal irony. Our main research concern was to provide evidence as to whether all cases of verbal irony are ‘critical’ in nature or not, both by means of theoretical reflection and by presenting the results of a survey whose questions were mainly oriented toward the identification of ironical situations by English native speakers. For our analysis of the evaluative content of ironic utterances we draw on the findings of studies such as Thompson and Hunston (2000), Martin and White (2005), or Bednarek (2008a). The results of the survey show that the great majority of speakers recognized the different evaluative shades of the ironical situations presented, which gave grounds for accepting our hypothesis that the essence of verbal irony is not to be found in its possible implied criticism, but in a finer, more subtle aspect of the phenomenon which has to do with a clash or contradiction at different linguistic/discursive levels.
FunGramKB (FGKB), on the one hand, is a multipurpose lexico-conceptual knowledge base for natural language processing (NLP) systems and comprises three major interrelated knowledge level modules: lexical, grammatical and conceptual. At the conceptual level the core ontology is presented as a hierarchical catalogue of the concepts that a person has in mind and a repository where semantic knowledge is stored. Axiology, on the other hand, is widely considered to be a primitive, basic or key parameter, among others, in the architecture of meaning construction at different levels. This parameter can be traced back to the three subontologies into which FunGramKB can be split: #ENTITY for nouns, # EVENT for verbs, and #QUALITY for adjectives. Even if most of the specific research conducted so far has been devoted to the category #QUALITY, there is no reason to consider verbs as less of an axiological category. Consequently, in this chapter we shall concentrate on the subontology # EVENT and explore how the main categories and features of the axiological parameter (good-bad or positive-negative [+/−]) are represented and encoded within FunGramKB ontology, particularly inside semantic properties such as basic or terminal concepts and meaning postulates, or syntactic operators, such as modality or polarity.
This chapter aims to examine the evaluative function of situation-bound utterances in intercultural interactions. The subject of inquiry is a unique formula whose use is tied to certain reoccurring situations. Situation-bound utterances (SBU) are frequently used in any language because these expressions serve as interactional patterns and rituals that usually mean the same to all speakers of a particular speech community. In a way, SBUs are reflections of the way native speakers of a language think. But what happens if the users of SBUs belong to different speech communities having a variety of L1s other than English? Will SBUs still keep their evaluative function (if any) when used in intercultural communication? These questions are very important if we accept that evaluative functions of language are culture- and language-specific. In order to answer the questions the chapter sets out to discuss the role of context in which situation-bound utterances are used.
The goal of this paper is to contribute to a better understanding of how prosodic lengthening in Spanish can be a means of marking information structure, particularly when an entire propositional content is involved. We test empirically the correlation between a form of prosodic marking used to identify a piece of content as already introduced in discourse – verum focus being a particular case of this general distinction – and evaluative values such as insistence and impatience. Native speakers consistently identify, in a perceptual experiment, the lengthened pattern with those discourse situations that involve repetition, often associated with insistence and impatience, and identify the nuclear stress as the locus of the marking.
This chapter investigates evaluation with respect to the hearer and it shows how evaluation permeates the phonological level of linguistic description. In particular, it examines how interlocutors of different cultural backgrounds, namely, English and Spanish, perceive and evaluate certain utterances as polite or rude based on their pitch range. A perception test was designed by means of which 15 native hearers of each language evaluated different productions of the word mandarins/mandarinas which only differed with respect to their F0 scaling. The results confirmed that pitch range differences have an effect on the evaluation of sentences cross-linguistically. Whereas low pitch range productions were interpreted as “polite” in Spanish and as “rude” in English, utterances with a high pitch displacement were judged as “over-excited” in Spanish and as “polite” in English. The results corroborate the status of intonation as an off-record strategy used to signal attitude.
In this chapter I explore evaluation in DVD blurbs (the text on the back of DVD box sets for TV series), which promote television series to potential buyers. The approach taken to the analysis of evaluation is corpus linguistics, and this chapter reports on findings from a pilot study examining the blurbs for 50 contemporary American television series. A combination of frequency analysis and concordancing is used to examine the use of evaluation in these discourse types in terms of frequency, dispersion and functionality. The findings suggest that the discourse type of the DVD blurb is linguistically restricted – although not in terms of specific lexico-grammatical items, but rather in terms of pragmatic functions, and that it has much in common with both traditional advertising and book blurbs.
This chapter reports research on evaluative language in English and Spanish consumer-generated reviews on books and movies. Within the Appraisal framework, a contrastive study is carried out on the spans of Graduation embedded in spans of Attitude in 64 reviews. A qualitative analysis, covering a syntactic description of these expressions of Graduation and proposals for doubtful cases, is followed by a quantitative analysis based on the parameters of language, subtypes of Graduation, product evaluated and positive or negative evaluation. The results show that the overall distribution of the main subtypes of Graduation spans in the English and the Spanish reviews is similar, but the consideration of each parameter in detail uncovers a number of differences between the reviews written in the two languages. In particular, the results point to possible pervasive differences in the expression of Affect in English and Spanish.
In this chapter, we investigate register diversification in evaluative language focusing on scientific writing. For this purpose, we explore selected epistemic and attitudinal variants of evaluative expressions. We consider (1) the situational context of these expressions according to Systemic Functional Linguistics and register theory (cf. Halliday and Hasan 1985; Biber 2006) and (2) selected aspects of their syntagmatic contexts (cf. Hunston 2011). The research questions we pose are (a) which typical evaluative expressions occur in scientific research articles and (b) whether the usage of particular evaluative expressions is rather discipline-specific or discipline-neutral. To answer these questions, we carry out a corpus-based analysis on a corpus of scientific research articles from nine scientific disciplines, combining quantitative and qualitative analytic methods.
This chapter explores the role of negative-modal synergies in the expression of authorial stance and intersubjective positioning in Charles Darwin’s The Origin of Species. As markers of stance, both negation and modality contribute to the expression of evaluation in discourse, though little attention has been paid to the co-occurrence of both types of markers. Drawing on corpus-based methods, I first identify the recurrent discourse pattern which gives rise to a semantic prosody of negative-modal meaning throughout The Origin of Species as compared to Voyage of the Beagle. Second, I discuss how this discourse pattern reflects Darwin’s positioning in the presentation of his Theory of Natural Selection. An analysis of the resources which express intersubjective positioning reveals the tension between conflicting goals in Darwin’s presentation of his new theory, namely, the expression of certainty regarding his insights and discoveries and the need to be cautious in communicating them. Thus, the various patterns of (co)-occurrence of negation, modality and personal pronouns construe specific authorial positions against the backdrop of competing scientific theories and in anticipation of readers’ potential disagreement.
This study uses a functional linguistic framework, the system of ENGAGEMENT (Martin and White 2005), to explore academic argumentation in course-related blogs from the standpoint of the ways in which blog writers make space for new knowledge against the background of other knowers’ ideas and ways they support their claims with evidence. The results reveal a significant presence of strategies typical of academic reasoning and suggest that it is a pattern of alternation between expanding and contracting options that lies behind more successful arguments. These findings tentatively confirm the conduciveness of the blog medium for practicing academic argumentation and point out the significance of the ENGAGEMENT framework as an analytical approach to academic argumentation and its usefulness for scaffolding learner argumentative writing.
In a world dominated by the Internet, it is becoming increasingly important to develop analytical tools that can take in multiple dimensions of media texts. This chapter presents a multimodal analysis of a corpus of online newspaper texts about controversies surrounding the wearing of religious items by Muslims, Sikhs and Christians, which received considerable media coverage in Britain in early 2010. Although the text itself generally provides a balanced account of the issues, a comparison of the results across modes shows that the groups are evaluated differently in visual images, headlines and direct quotations. This study represents a step on the road to devising an integrative model for multimodal analysis.
This chapter investigates the expression of evaluation in a corpus collected from weekly news magazines in English in an attempt to address two main research questions. I investigate, first, the linguistic patterns regularly associated with the expression of evaluation in this written register and, second, the place of evaluative expressions in this discourse type and the implications for the organization of the text. The analysis has identified five main patterns associated with evaluation: copular constructions, cleft constructions, lexical means, markers of modality and other devices, such as conditional or comparative clauses in contexts of evaluation. For the second question, evaluation has been found to concentrate in some texts either at the end or at a point where it reflects the views of a third party in the story. However, there is insufficient evidence to claim that in news magazines evaluation tends to occur at boundary points in the text.
This chapter continues investigation into register-idiosyncratic features of evaluation in parliamentary debate (Miller 2007; Miller and Johnson 2009, to appear), reporting findings regarding the phraseology it is * time to/for/that in a corpus of US congressional speech on the Iraq war. Quantitative data are tested for saliency against both general and political corpora. Qualitative analysis focuses on the enactment of APPRAISAL systems (Martin and White 2005). Methodology involves “shunting” (Halliday 2002 [1961]: 45), and a focus on “coupling” (Martin 2000: 163–164) which affords attitude (Martin and White 2005: 62 ff.; Miller and Johnson, to appear). The chapter examines how appraisers’ choices are affected by party/gender, recognizing that choice may transcend register boundaries due to both the ‘repertoire’ of the individual and his/her ideologically saturated ‘reservoir’ of culturally specific ways of meaning (Martin 2010: 23).
Narratives are cognitive means of organizing and constructing our experience in a particular way. In emotion narratives, narrators do more than report propositional information; they need to attract the listener’s empathy and understanding. By applying sociocognitive and functional models of language and discourse (Herman 2003; Redeker 2006; Bernárdez 2008) in order to complement the Labovian approach to narratives, this study shows (i) that, in this specific text type, evaluation – the expression of the narrator’s emotions – is not an independent section, but suffuses the whole texts from beginning to end, and (ii) that the recurrent linguistic and pragmatic strategies chosen in order to disclose highly personal information – discourse markers, repetitions, repairs, profusion of details, etc. – are related to the specific linguistic activity and discourse context.
This chapter explores evaluative discourse on social networking sites (henceforth SNSs), paying special attention to the role played by the expression of attitude and positive politeness in the management of interpersonal rapport. The corpus for the study consists of a random sample of 100 messages exchanged among university students in the United Kingdom and the United States on a particular site, i.e., Facebook, during the two-year period 2010–2012. Analysis is approached from the theories of appraisal (Martin and White 2005; Bednarek 2008) and politeness (Brown and Levinson 1987); and the methodology for processing the data borrows quantitative techniques from Corpus Linguistics. The findings indicate that specific contextual features of SNSs seem to trigger the production of attitudinal meanings of affect, judgement and appreciation, which are exploited for the relational work involved in the construction and maintenance of positive face.
This article sets out to provide a summarized historical survey of some relevant contributions to axiological semantics from linguists included in the two most influential linguistic traditions from the 50s to the late 70s: structural linguistics and transformational-generative grammar. In fact, even if their representative opinions at that time could now be regarded as incidental, it would be unfair to claim that some of their intuitions and principles have not been crucial in illuminating essential aspects of the parameter of axiological evaluation in modern linguistics. In this respect, it is worth highlighting the ideas disseminated by key figures of the Saussurean or Chomskyan traditions from the perspective given by half a century of evolution in the field. For that purpose contributions are included from authors such as Apresjan, Bally, Coseriu, Ducháček, Fodor, Grzegorek, Guiraud, Katz, Klima, Pottier, Stati, and Ullmann, among others.
The focus in this chapter is on the evaluative systems of attitude, as modelled in the appraisal framework. While fully accepting the validity of the model, I explore three practical issues that arise in applying it. The first concerns the question of whether affect, the core area of attitude, should be distinguished from the more general field of emotion talk, and, if so, how. The second issue arises because two related but distinct criteria are used to distinguish judgements and appreciations: the kind of entity evaluated (the target) and the value ascribed to it. As a result, there may be an apparent mismatch between these two criteria. The third difficulty can be called the ‘Russian doll’ dilemma. An expression of one category of attitude may function as a token (an indirect expression) of a different category; and that token may itself function as an indirect expression of yet another category, and so on. This poses problems for a study of evaluation which relies on quantitative analysis. I suggest my own principles to help make life a little easier for anyone engaged in analysing evaluation in text.
Appraisal is a term referring to systems within Systemic Functional Linguistics that map evaluative language. This chapter reviews two Appraisal systems (Attitude and Graduation) and assesses their viability for analysing evaluation in narratives and students’ written responses to these. Within this context, a crucial affordance of Appraisal is its semantic orientation, which allows analysts to capture the delicate interweaving of explicit and implicit attitudes that powerfully positions readers of narratives. However, the sprawling, cumulative and context-sensitive nature of evaluation requires a rigorous methodology for using Appraisal for coding purposes. We address this challenge, proposing a cline of implicitness that relates degrees of reliability to instances of evaluation depending on their textual and/or cultural/institutional environments.
In this chapter we present the results of both a theoretical and empirical study on the evaluative character of verbal irony. Our main research concern was to provide evidence as to whether all cases of verbal irony are ‘critical’ in nature or not, both by means of theoretical reflection and by presenting the results of a survey whose questions were mainly oriented toward the identification of ironical situations by English native speakers. For our analysis of the evaluative content of ironic utterances we draw on the findings of studies such as Thompson and Hunston (2000), Martin and White (2005), or Bednarek (2008a). The results of the survey show that the great majority of speakers recognized the different evaluative shades of the ironical situations presented, which gave grounds for accepting our hypothesis that the essence of verbal irony is not to be found in its possible implied criticism, but in a finer, more subtle aspect of the phenomenon which has to do with a clash or contradiction at different linguistic/discursive levels.
FunGramKB (FGKB), on the one hand, is a multipurpose lexico-conceptual knowledge base for natural language processing (NLP) systems and comprises three major interrelated knowledge level modules: lexical, grammatical and conceptual. At the conceptual level the core ontology is presented as a hierarchical catalogue of the concepts that a person has in mind and a repository where semantic knowledge is stored. Axiology, on the other hand, is widely considered to be a primitive, basic or key parameter, among others, in the architecture of meaning construction at different levels. This parameter can be traced back to the three subontologies into which FunGramKB can be split: #ENTITY for nouns, # EVENT for verbs, and #QUALITY for adjectives. Even if most of the specific research conducted so far has been devoted to the category #QUALITY, there is no reason to consider verbs as less of an axiological category. Consequently, in this chapter we shall concentrate on the subontology # EVENT and explore how the main categories and features of the axiological parameter (good-bad or positive-negative [+/−]) are represented and encoded within FunGramKB ontology, particularly inside semantic properties such as basic or terminal concepts and meaning postulates, or syntactic operators, such as modality or polarity.
This chapter aims to examine the evaluative function of situation-bound utterances in intercultural interactions. The subject of inquiry is a unique formula whose use is tied to certain reoccurring situations. Situation-bound utterances (SBU) are frequently used in any language because these expressions serve as interactional patterns and rituals that usually mean the same to all speakers of a particular speech community. In a way, SBUs are reflections of the way native speakers of a language think. But what happens if the users of SBUs belong to different speech communities having a variety of L1s other than English? Will SBUs still keep their evaluative function (if any) when used in intercultural communication? These questions are very important if we accept that evaluative functions of language are culture- and language-specific. In order to answer the questions the chapter sets out to discuss the role of context in which situation-bound utterances are used.
The goal of this paper is to contribute to a better understanding of how prosodic lengthening in Spanish can be a means of marking information structure, particularly when an entire propositional content is involved. We test empirically the correlation between a form of prosodic marking used to identify a piece of content as already introduced in discourse – verum focus being a particular case of this general distinction – and evaluative values such as insistence and impatience. Native speakers consistently identify, in a perceptual experiment, the lengthened pattern with those discourse situations that involve repetition, often associated with insistence and impatience, and identify the nuclear stress as the locus of the marking.
This chapter investigates evaluation with respect to the hearer and it shows how evaluation permeates the phonological level of linguistic description. In particular, it examines how interlocutors of different cultural backgrounds, namely, English and Spanish, perceive and evaluate certain utterances as polite or rude based on their pitch range. A perception test was designed by means of which 15 native hearers of each language evaluated different productions of the word mandarins/mandarinas which only differed with respect to their F0 scaling. The results confirmed that pitch range differences have an effect on the evaluation of sentences cross-linguistically. Whereas low pitch range productions were interpreted as “polite” in Spanish and as “rude” in English, utterances with a high pitch displacement were judged as “over-excited” in Spanish and as “polite” in English. The results corroborate the status of intonation as an off-record strategy used to signal attitude.
In this chapter I explore evaluation in DVD blurbs (the text on the back of DVD box sets for TV series), which promote television series to potential buyers. The approach taken to the analysis of evaluation is corpus linguistics, and this chapter reports on findings from a pilot study examining the blurbs for 50 contemporary American television series. A combination of frequency analysis and concordancing is used to examine the use of evaluation in these discourse types in terms of frequency, dispersion and functionality. The findings suggest that the discourse type of the DVD blurb is linguistically restricted – although not in terms of specific lexico-grammatical items, but rather in terms of pragmatic functions, and that it has much in common with both traditional advertising and book blurbs.
This chapter reports research on evaluative language in English and Spanish consumer-generated reviews on books and movies. Within the Appraisal framework, a contrastive study is carried out on the spans of Graduation embedded in spans of Attitude in 64 reviews. A qualitative analysis, covering a syntactic description of these expressions of Graduation and proposals for doubtful cases, is followed by a quantitative analysis based on the parameters of language, subtypes of Graduation, product evaluated and positive or negative evaluation. The results show that the overall distribution of the main subtypes of Graduation spans in the English and the Spanish reviews is similar, but the consideration of each parameter in detail uncovers a number of differences between the reviews written in the two languages. In particular, the results point to possible pervasive differences in the expression of Affect in English and Spanish.
In this chapter, we investigate register diversification in evaluative language focusing on scientific writing. For this purpose, we explore selected epistemic and attitudinal variants of evaluative expressions. We consider (1) the situational context of these expressions according to Systemic Functional Linguistics and register theory (cf. Halliday and Hasan 1985; Biber 2006) and (2) selected aspects of their syntagmatic contexts (cf. Hunston 2011). The research questions we pose are (a) which typical evaluative expressions occur in scientific research articles and (b) whether the usage of particular evaluative expressions is rather discipline-specific or discipline-neutral. To answer these questions, we carry out a corpus-based analysis on a corpus of scientific research articles from nine scientific disciplines, combining quantitative and qualitative analytic methods.
This chapter explores the role of negative-modal synergies in the expression of authorial stance and intersubjective positioning in Charles Darwin’s The Origin of Species. As markers of stance, both negation and modality contribute to the expression of evaluation in discourse, though little attention has been paid to the co-occurrence of both types of markers. Drawing on corpus-based methods, I first identify the recurrent discourse pattern which gives rise to a semantic prosody of negative-modal meaning throughout The Origin of Species as compared to Voyage of the Beagle. Second, I discuss how this discourse pattern reflects Darwin’s positioning in the presentation of his Theory of Natural Selection. An analysis of the resources which express intersubjective positioning reveals the tension between conflicting goals in Darwin’s presentation of his new theory, namely, the expression of certainty regarding his insights and discoveries and the need to be cautious in communicating them. Thus, the various patterns of (co)-occurrence of negation, modality and personal pronouns construe specific authorial positions against the backdrop of competing scientific theories and in anticipation of readers’ potential disagreement.
This study uses a functional linguistic framework, the system of ENGAGEMENT (Martin and White 2005), to explore academic argumentation in course-related blogs from the standpoint of the ways in which blog writers make space for new knowledge against the background of other knowers’ ideas and ways they support their claims with evidence. The results reveal a significant presence of strategies typical of academic reasoning and suggest that it is a pattern of alternation between expanding and contracting options that lies behind more successful arguments. These findings tentatively confirm the conduciveness of the blog medium for practicing academic argumentation and point out the significance of the ENGAGEMENT framework as an analytical approach to academic argumentation and its usefulness for scaffolding learner argumentative writing.
In a world dominated by the Internet, it is becoming increasingly important to develop analytical tools that can take in multiple dimensions of media texts. This chapter presents a multimodal analysis of a corpus of online newspaper texts about controversies surrounding the wearing of religious items by Muslims, Sikhs and Christians, which received considerable media coverage in Britain in early 2010. Although the text itself generally provides a balanced account of the issues, a comparison of the results across modes shows that the groups are evaluated differently in visual images, headlines and direct quotations. This study represents a step on the road to devising an integrative model for multimodal analysis.
This chapter investigates the expression of evaluation in a corpus collected from weekly news magazines in English in an attempt to address two main research questions. I investigate, first, the linguistic patterns regularly associated with the expression of evaluation in this written register and, second, the place of evaluative expressions in this discourse type and the implications for the organization of the text. The analysis has identified five main patterns associated with evaluation: copular constructions, cleft constructions, lexical means, markers of modality and other devices, such as conditional or comparative clauses in contexts of evaluation. For the second question, evaluation has been found to concentrate in some texts either at the end or at a point where it reflects the views of a third party in the story. However, there is insufficient evidence to claim that in news magazines evaluation tends to occur at boundary points in the text.
This chapter continues investigation into register-idiosyncratic features of evaluation in parliamentary debate (Miller 2007; Miller and Johnson 2009, to appear), reporting findings regarding the phraseology it is * time to/for/that in a corpus of US congressional speech on the Iraq war. Quantitative data are tested for saliency against both general and political corpora. Qualitative analysis focuses on the enactment of APPRAISAL systems (Martin and White 2005). Methodology involves “shunting” (Halliday 2002 [1961]: 45), and a focus on “coupling” (Martin 2000: 163–164) which affords attitude (Martin and White 2005: 62 ff.; Miller and Johnson, to appear). The chapter examines how appraisers’ choices are affected by party/gender, recognizing that choice may transcend register boundaries due to both the ‘repertoire’ of the individual and his/her ideologically saturated ‘reservoir’ of culturally specific ways of meaning (Martin 2010: 23).
Narratives are cognitive means of organizing and constructing our experience in a particular way. In emotion narratives, narrators do more than report propositional information; they need to attract the listener’s empathy and understanding. By applying sociocognitive and functional models of language and discourse (Herman 2003; Redeker 2006; Bernárdez 2008) in order to complement the Labovian approach to narratives, this study shows (i) that, in this specific text type, evaluation – the expression of the narrator’s emotions – is not an independent section, but suffuses the whole texts from beginning to end, and (ii) that the recurrent linguistic and pragmatic strategies chosen in order to disclose highly personal information – discourse markers, repetitions, repairs, profusion of details, etc. – are related to the specific linguistic activity and discourse context.
This chapter explores evaluative discourse on social networking sites (henceforth SNSs), paying special attention to the role played by the expression of attitude and positive politeness in the management of interpersonal rapport. The corpus for the study consists of a random sample of 100 messages exchanged among university students in the United Kingdom and the United States on a particular site, i.e., Facebook, during the two-year period 2010–2012. Analysis is approached from the theories of appraisal (Martin and White 2005; Bednarek 2008) and politeness (Brown and Levinson 1987); and the methodology for processing the data borrows quantitative techniques from Corpus Linguistics. The findings indicate that specific contextual features of SNSs seem to trigger the production of attitudinal meanings of affect, judgement and appreciation, which are exploited for the relational work involved in the construction and maintenance of positive face.