Analytical Forays into Functionalism: Theory and Practice

Abstract

Presentation

1. Presentation: The proposal for a dossier of functionally oriented studies

In the interests of strengthening research links between Brazilian and foreign universities, Abralin (the Brazilian Association of Linguistics) decided to promote the production of four special thematic dossiers on Linguistics, with researchers from Brazil and other countries, for publication in its Revista da Abralin, including texts not only in Portuguese but also in English.

In historical terms, it is significant that this Abralin initiative came immediately after the Association’s promotion of “Abralin ao Vivo”, a series of events (lectures and round tables) which took place online from May 2020 onwards, and which is extending into 2021. Abralin justified this project by reflecting on the times we were living through, marked by the social distancing that the pandemic has imposed. At the same time it reminded its members of the need to preserve “the advancement of science and the exchange of ideas among researchers” in order “to continue supporting technical interventions and political decisions based on scientific knowledge”. It was in this way that various research dissemination activities took off that involved – and continue to involve, and bring together – a large number of researchers from Brazil and beyond, including resounding names from the science of linguistics, including, relevantly, research that is functionalist in orientation.

This Abralin initiative deserves to be highlighted as having great importance for the macrocosm of linguistics, indeed as a milestone in the internationalization of linguistic research, particularly in terms of disseminating work from the Brazilian research context. It is worth mentioning that non-Brazilian colleagues have stated that few researchers from other countries knew Abralin before the year 2020 but that “Abralin ao Vivo” made a deep and very positive impression on everyone who participated in its events, as lecturers, speakers, moderators or listeners. And it is important to note that, of the authors included in this dossier, four participated in that event, either as lecturers or as members of thematic panels. It was unfortunately impossible to involve other Abralin ao Vivo participants in this special issue of the journal, since the schedule for the submission of manuscripts was necessarily very tight. Two of the organizers of this dossier, Maria Helena de Moura Neves and Lachlan Mackenzie, were also speakers at the event, but the editors considered it inappropriate to compete for a place in this work.

2. The nature and general set-up of the dossier

After the launch of Abralin’s call, we three submitted a proposal to the Presidency of the Association, offering Linguistic Functionalism, or Functional Linguistics, or Functional Grammar, as a thematic focus. Our proposal aimed at producing a special issue of the journal (a ‘dossier’) that would map out selected contemporary themes from the broad spectrum of functionalist theoretical orientations. The general direction of the proposal thus favored an emphasis on discussing theoretical matters rather than expounding existing functionalist approaches.

The summary sent to Abralin laid out the following guidelines for this issue of Revista da Abralin:

1) In Linguistic Functionalism, language is conceived of as an evolutionary-adaptive phenomenon through which the grammars of natural languages ​​are organized on the basis of descriptions of linguistic structure.

2) It is held that, in this context, it is possible to reach explanations for the ways in which the grammars of natural languages ​​yield descriptions of linguistic structure, with due consideration of the interfaces among the syntactic, semantic and pragmatic components.

3) This type of description is effectively underpinned by countless functionalist proposals that give pride of place to the semantico-pragmatic organization of grammar in order to explain the grammatical processes that constitute utterances, with a view either to the socio-cognitive component or to typological and universal aspects.

The proposal went on to indicate that, within this framework, the editors of the dossier intended to bring together studies that were representative of the contemporary development of various strands of functionalism. This broad orientation embraced the description of grammatical aspects of natural languages as well as studies that discuss, in theory and in practice, such themes as: the socio-cognitive framework of language use; the relationship between speech and grammar; iconicity, economy and competing motivations; variation and change in the grammar of natural languages; mappings between cognitive and linguistic categories; the fluidity of categories; prototypes and exemplars; the grammatical processes that underlie utterances; the relationship among grammatical constructions; linguistic typology; linguistic universals. Related topics, treated from either a theoretical or an applied point of view, would also be considered for the dossier.

As for the profile of the authors we wished to attract, it was understood that we would not necessarily privilege experienced scholars but would also welcome young researchers, the decisive factor being the representativeness of the material to be published.

3. A meta-analysis of the eight functionalist forays that make up the dossier

The dossier opens with Adriana Pagano’s article Modeling Language and Context in Systemic-Functional Theory, which offers a very careful account of Halliday’s systemic-functional architecture, incorporating the updates that the theory has undergone. The account delineates the dimensions that this particular architecture proposes for a general understanding of the organization of language: the “global dimensions” (stratification, instantiation and metafunction), and the “local dimensions” (structure and system). The article’s route map also takes in Halliday’s two central ideas: that language is “naturally” complex; and that language resolves theoretically into “choices”, implying “selections”. This brings out the relevance of a paradigmatic approach to the availability of options within what is encompassed by the different dimensions of language organization. The corollary is that meaning is generated by systemic contrast, by what arises between each form that was chosen and all other forms that failed to be chosen. And it is this notion of choice that is also involved in the systemic-functional concept of “metafunctions”, definable as co-occurrences of systemic selections dedicated not only to the construction of our experience of the world and our social relations but also to the construction of text itself – an enabling and facilitating function of language, therefore.

The next article, Efficient Trade-Offs as Explanations in Functional Linguistics: Some Problems and an Alternative Proposal, by Natalia Levshina, takes data from a corpus of 30 languages (the Universal Dependencies database) and an annotated corpus of online news (from the Leipzig Corpora collection) to discuss the notion of an “efficient trade-off”, a concept currently invoked in functional linguistics to explain various aspects of language use and structure. In defining efficiency with reference to a cost-benefit calculation, linguists have applied this concept of an “efficient trade-off” to negative correlations between two explanations that run in opposite directions, each of them resulting from “rational choices”. The article argues that there is no benefit to taking this intuitively attractive notion as a guide for analysis, considering it “more confusing than enlightening”, not least because not every negative correlation represents a real trade-off. The author adduces numerous problems with the use of the concept of “efficient trade-off” for explanations in functional linguistics, pointing out the advantages of replacing analyses that are based on negative correlations between pairs of linguistic variables with a causal analysis that covers a multitude of factors and problematizes the consensus idea that languages ​​tend to be efficient. The upshot is that it would be appropriate to completely abandon the treacherous concept of “efficient trade-offs”, replacing the use of correlations with the identification of unidirectional causal relations, exemplified in the article by the conclusion drawn by the author from her analysis of data that word order determines case-marking, but not vice versa.

The third article, A Functionalist Typology of Redundancy by Sterre Cécile Leufkens, is dedicated to the study of linguistic redundancy – the “repetition of information”, as defined by the author – a grammatical phenomenon that has been highlighted as being of particular interest in functionalist research. It has been pointed out that this phenomenon, which is very common in colloquial language, is fundamental to the architecture of language (and even may be a linguistic universal) but has received little empirical study, perhaps due to the fact that, since redundancy phenomena are “omnipresent” in languages, they appear to be impossible to study. And it is precisely this – according to the author – that makes syntagmatic redundancy into an object of research that is particularly relevant from the functionalist perspective, as this very omnipresence is easily attributable to its functionality. The objective of Leufkens’ study is presented as a functional search, conducted through the analysis of reference grammars from a variety sample of 50 languages, for the motivations behind the different forms of redundancy, as well as the communicative functions of four types of concord selected for examination. The results throw up two valuable answers: certain types of agreement are shown to be almost universal while others are restricted to a few languages; and it is the need for precision and the need for emphasis on crucial information that are the factors that motivate concord, the former yielding “accidental” and the latter “purposeful” redundancy. In addition to these conclusions, the article establishes that, for each of these two types of redundancy, there is a characteristic diachronic development.

The article Typology and Nuance: Relativization by Marianne Mithun defends the idea that work in linguistic typology is extremely valuable for exploring not only what languages ​​have in common but also how they differ from one another, as well as the possible correlations among these variables. The assumption has been that linguistic research is primarily concerned with the elements shared by languages, moving on to evaluate how they vary and to determine the causes of that variation through the identification of categories that underpin such comparative work. The starting point of Mithun’s argument is that most of what has been offered in linguistic typology is inherently categorical, i.e., classifications are made in terms of the “presence” or “absence” of features, a choice that in certain cases permits well-founded analyses, but not in many others that require the adoption of more attentive consideration of the nuances of language. In her view, what is lacking is an assessment of the extent to which a particular linguistic expression, or a specific distinction, or a variable, is categorically delimited and decisively established in the grammar of the language. As for the methodology of research, the author proposes that work should begin with the transcription of spontaneous speech, thus guaranteeing analysis that includes the functionality of the context of utterance, accompanied by prosodic analysis. The phenomena analyzed in the article are from Mohawk, an indigenous Iroquois language of Northeastern North America, and concern the construction of “relative” clauses, the hypothesis being that these constructions “do not seem to be well installed in grammar”. The author considers that the results of her analysis, revealing particular properties of that language, are valuable for typological research and, moreover, can be generalized to functionally oriented descriptive work in ways that deserve to be highlighted. A particularly relevant aspect of this article is its demonstration that work in linguistic typology can be enriched through the recognition that the “categories” and “variables” that make up theories are not always rigidly “categorical” in particular languages.

The fifth article, The Functional Spectrum of Bem and Bom in Spoken Portuguese: Instances of Grammaticalization by Edair Maria Gorski, provides a functional description of the items bem ‘well’ and bom ‘good’. Its point of departure is the fact, already well-established in previous research, that these words, categorized with regard to their origins as an adverb (from the Latin bene) and an adjective (from the Latin bonus, a, um) respectively, present both uses in which the original categories are preserved and uses with the acquired function of discourse markers, in addition to yet others. Examining dialogical situations in 72 sociolinguistic interviews from the VARSUL Project, the author provides a comparative account of the discourse uses of the items under examination, with the purpose of understanding behavioral patterns. She considers the items to be instances of “grammaticalization” (or, more strictly, of “expansion”), interpreting the changes she has observed and highlighting the role of context. With an emphasis on the fact that grammar involves, in addition to a strictly structural level, a semantico-pragmatic level, the author regards discourse markers as belonging to grammar, since they fulfil grammatical functions. The focus of her investigation is on the performance of bem and bom as conversational markers; all the functions found in the data are described and discussed, yielding, with regard to their occurrence in context, a more accurate view of these particular items. In general, the results point to functional differences in the use of each (in their various roles), but it deserves to be stressed that when they function as discourse markers, the two items share, within their class, one general (textual/interpersonal) property: as the author notes, it is that of “signaling a double movement (of discursive structuring and negotiation), with a significant presence at the head of immediate responses”. In any case, as is concluded in the study, the result of the analysis will depend on the communicative situation of the discourse genre in which they are operative.

The following article, Extra-Clausal Constituents in Functional Discourse Grammar: Function and Form by Riccardo Giomi and Evelien Keizer, develops what they call “a first outline of an encompassing account of extra-clausal constituents (ECCs) within the framework of Functional Discourse Grammar (FDG)”. The authors’ stance differs from those studies on grammaticalization and pragmaticalization that regard these elements, which are so directly linked to discourse functions, as being “external” to the grammar itself; it also differs from those that remove these elements from a “sentence grammar” (i.e. from intraclausal relations) and put them in a “thetic grammar” (of extraclausal relations). Employing the theoretical apparatus of FDG, the authors argue that what determines which expressions belong to the grammatical component is not a question of “interiority”, or not, of the expressions in relation to the clause, but rather the question of the “interiority”, or not, of their semantic and pragmatic content in the structure of the language. As they say, it is matter of there being “explicit, systematic encoding in language structure”. In the theoretical framework of FDG, everything that a speaker expresses morphosyntactically and/or phonologically and which fulfils a communicative intention has grammatical status. Noting that FDG does not yet have ways of determining precisely what the definition is of an “extra-clausal constituent”, the study offers an investigation of the functional and formal properties of the elements of this class (however heterogeneous it may be). It establishes, as central points of observation and definition, that ECCs share a common underlying pragmatic structure reflecting their contribution to the development of discourse and investigates the communicative factors that determine the significant occurrence of these items in language. It is proposed, specifically, that “each ECC forms a separate though invariably dependent unit of communicative behaviour (i.e. a Subsidiary Discourse Act)”, i.e., not a component of the expression that precedes, follows or interrupts it. On this basis, the authors move on to consider the mechanism that governs the positioning of ECCs, in relation to the communicative units to which they are added, in ways that differ from current proposals in FDG. The authors conclude that their study has made it possible to define functional restrictions on the positioning of ECCs and to account for the greater degree of positional freedom that these constituents enjoy in relation to those that belong to lower layers of morphosyntactic organization.

In the following article, authors Marta Anaisa Bezerra Ramos and Camilo Rosa consider a question that is particularly interesting for functional analysis, namely the nature of adverbial clauses of manner in Portuguese. The title, Adverbial Clauses of Manner Introduced by sem/sem que ‘without’: One Look at the Grammatical Tradition and Another at Their Uses, already provides a condensed indication of the object of the authors’ research; their argument, as a whole, takes a direct route to the matter under analysis, which gives great unity to the text. This is a phenomenon that has unusual properties, with regard to both the categorical status of the connectors and the semantic range of the resulting adverbial constructions. It is worth noting that these clauses are not classified in the Brazilian grammatical nomenclature and are rarely treated in grammatical works in general. The article is founded on a functionalist theoretical basis that conceives grammar as being related to discourse as well as cognition. In the analysis, the authors set up criteria for recognizing manner relations indicated by sem or sem que and for distinguishing them from other adverbial relations, a challenging task given the great fluctuation in the meanings involved. Their study of clauses introduced by sem and sem que is based on a corpus of opinion articles, editorials and interviews from Brazilian weekly newspapers. The analysis reveals a high frequency of a manner value for these structures, allowing the authors to achieve the objective that was already central to their initial proposal, namely to demonstrate that there is a particular domain of ‘manner’, not a broad domain of various adverbial semantic relations such as comparison, conformity or concession in which this particular value would be diluted.

The final article of this dossier, Reacting to Social Quotas: A Study of Facebook Comments in Brazilian Portuguese, authored by Rodrigo Esteves de Lima Lopes, is relevant not only for functionalist studies, and specifically for Systemic-Functional Grammar, but also for studies in corpus linguistics. The abstract states, in a very practical initial formulation, that the objective of the paper is “to discuss how Facebook users reacted to the news that the Federal University of Bahia (Brazil) had introduced social quotas for transsexuals, immigrants and socially vulnerable refugees”. The text opens with information about how, for the purposes of grammatical description, the study is anchored in a corpus and how the analysis will be oriented to lexicogrammar. Being embedded in the systemic-functional apparatus, the study’s main focus is on establishing the patterns of “transitivity” and “evaluation” represented by the social actors. The methods of collecting and processing the data are carefully specified, and the approach is supported by computational statistical analysis. The results indicate, in general terms, that most of the comments under examination come from the interaction between Facebook users and that these comments, not always friendly, occur in discussions that concern the positions taken by the speakers. One particularly relevant indication is that the discussions are politically polarized, typically going beyond a simple expression of opinions that are favorable or opposed to the quota system. Another is that there are frequent manifestations of prejudice against people from the Northeast of Brazil and against the very states they come from. An important aspect of the study is that the theoretical framework permits one to identify grammatical patterns for each of the themes present in the comments. Using the systemic-functional approach to analysis, the author was able to establish, by examining the meanings found in the comments under analysis, the systems of choice relating to evaluation and transitivity.

4. The aftermath of our proposal

Taking a global view, we observe that the articles that make up this dossier are organized into two groups with regard to the nature of their contents. The first three are driven by the problematization of some theoretical point: in Pagano’s text, the legitimacy of context as a theoretical category in systemic-functional linguistics, and, in the contributions by Levshina and Leufkens, the relevance of communicative pressures as explanatory categories in linguistic typology. The fourth article, by Mithun, problematizes traditional practice in linguistics and particularly in typological studies, that of seeing grammatical categories as inherently categorical, that is, as discrete. Developing this question theoretically, Mithun proposes a more nuanced view of grammatical categories by analysing the grammaticalization of relative clauses in Mohawk.

Mithun’s article serves as a transition between the first three articles, marked by discussion of theoretical principles, and the last four, which are oriented to the analysis of specific grammatical facts in particular languages: the grammaticalization of discourse markers in Portuguese (Gorski), the formal and functional properties of extraclausal constituents in English (Giomi; Keizer), the discursive distribution of adverbial manner clauses in Portuguese (Ramos; Rosa), and lexicogrammatical choices in the constitution of discourse on social quotas in Portuguese (Lima-Lopes).

All in all, the dossier gathers together studies that widen the reader’s field of view, laying bare the foundations of relevant functional theories and proposing new guidelines, alongside studies that draw the reader’s attention to new analyses, some focused, others more nuanced, but all supported by functionalist thinking.

We wish to end by thanking all the authors and the reviewers for their excellent contributions to the dossier and for uncomplainingly accepting the very strict and demanding deadlines that had to be imposed.