Discourse markers, also known as metadiscourse features (Hyland, 2018), theticals (Heine, 2013) and Pragmatic markers (Brinton, 2017), have been an ongoing topic of research by courtesy of a big body of research and an ensemble of trendsetter books in the last years. Despite the fact that the concept of discourse markers has gained currency; thanks to the large thread of research (see for example Brinton, 2011, 2017; Jucker, 1997 and Müller, 2005), in a bid to push the boundaries and in a nascent call for further study in this area of knowledge, the current volume, which is a timely contribution under the editorship of Cuenca and Degand (2022), has come to fruition.

Structurally, this book consists of 10 chapters revolving thematically around the concept of discourse markers. At the first impression and when read from cover to cover, it becomes conspicuous that the authors of this volume have been successful in their efforts to convene preeminent researchers to compile a trailblazing work of scholarship. The book at hand comes to offer state of the art perspectives and studies to the concept of discourse markers and is basically a concerted effort to approach them from different angles including experimental or corpus-based studies.

The first chapter, which is an introduction to the whole book, gives a general review of the concept of discourse markers. Chapter two reports an experimental approach to discourse marking which consists of a series of untimed eye-tracking experiments in English, Spanish and German with presence and absence of discourse markers. The results showcase that there are three cognitive principles of discourse marking which profoundly affect utterance processing: (1) The presence of discourse markers modifies the processing strategy in relation to an unmarked utterance; (2) the introduction of a discourse marker in a specific utterance sets as its maximum processing costs those of the corresponding unmarked utterance; and (3) the introduction of a discourse marker in a given utterance sets as the upper limit of processing efforts of the segment in which it is integrated.

The following chapter by Blochowski & Griost, through experimental research design, differentiates between two commonly used discourse markers: car and parce que. The impetus of running this study stems from the fact that evidence has demonstrated that car has gained ground in expressing objective relations despite the fact that it is basically used as a subjective relation discourse marker. To conduct this study, the researchers utilized two experiments: an offline experiment with university students and a crowdsourcing experiment with a general population. The results show that the use of two discourse markers depends, largely, on various factors such as the speaker’s educational background, age and reasoning capabilities. It is not stated, however, what is meant by offline and crowdsourcing experiments. It would have been more informative if the authors had explained the nature of these experiments.

The fourth chapter, written by Zufferey et al., zooms in on the ability of German-speaking learners of French to master connectives and discourse markers through a series of experiments including a fill-in-the-blanks test as well as a French vocabulary test among three groups of native adult speakers, native teenagers and second language learners. The results pinpointed that the frequency rank of connectives (discourse markers) cannot always account for learners’ predicaments in using connectives appropriately. In addition, the results showed that cognitive complexity may not be the major reason either, as in all three groups exposure to print and reading skills play a key role in predicting learners’ability to deal with connective markers as compared to verbal skills.

Chapter five, by Wetzel, Zufferey and Gygax, is an experimental research with the prime objective of comparing native and second language learners. The general hypothesis in this study is that information conveyed by connectives is less salient when reading in a second language than in a native language. The authors report two experiments of self-paced reading times of native and non-native readers of French for causal and concessive clauses. The results indicated that readers tend to read implicitly marked sentence more slowly than explicitly. In addition, it was found that reading in a second language mostly follows L1 cognitive principles and patterns given the fact that non-native readers read connective sentences more slowly when compared to native speakers.

Written by Broisson & Degand, chapter six is an immense effort to uncover the effects of speakers’ vs. hearers’ orientations (egocentric vs. allocentric) and cognitive load on the use of discourse markers. In a call for further study, they echo the claim that discourse markers have been so far labeled as reader- or hearer-oriented, which this research tries to put this claim to the test while also revealing the effects of cognitive load on their production. The data were elicited from 24 participants doing two speaking tasks with two variables being manipulated. The results showcased that discourse markers with a higher frequency will not necessarily exert effects on cognitive load. In the same manner, discourse markers were more frequent in the allocentric task than in the egocentric task. All in all, the results indicated that speakers prefer to use more polysemous discourse markers and take hearers’ processing cost into account in order to facilitate the comprehension process for them.

Chapter seven, written by Li et al., mirrors the results of an empirical study on the effect of discourse markers on computational sentence understanding and computational models for language processing. Indeed, this research focuses, exclusively, on evaluating how connectives are leveraged in sentence embedding models and tries to unpack the effects of discourse markers on sentence embedding computational models and how it differs from human sentence processing. The results pinpoint that discourse marker relation, such as sequential, adversative and additive, can potentially affect processing relations. In other words, omitting discourse markers can have negative impacts on computational language processing, whereas sequential discourse markers are easier to comprehend. Overall, the results indicate that presence and appropriate use of discourse markers will reduce processing time and accelerate reading continuity.

Chapter eight, by Verdonik, zooms in on the way discourse marker annotation can be applied in computational linguistics. The prime objective of this chapter is to reveal the interaction between discourse functions, on the one hand, and dialogue segmentation, on the other hand. Verdonik determines when a discourse marker can be used as a dialogue act on its own and when it becomes part of a dialogue act in Slovene informal dialogue. The corpus analysis demonstrates conclusively that there are six criteria which precisely determine dialogue acts of discourse markers including separation from the other discourse through pauses, emphasis in pronunciation, discourser markers as verbal imperatives in their grammatical structures and discourse markers reflecting an intention of starting an utterance but interrupted by an interlocutor.

Given the role of parallel corpora in translation research we have seen a rise in this strand of study (Vasheghani Farahani, 2022). In pursuit of further study, the ninth chapter of the book has been dedicated to translating discourse markers through utilizing parallel and directional corpora. In this regard, Cuenca strives to uncover translation strategies and factors in relation to discourse markers. By resorting to a parallel corpus of 10 papers in Catalan and English, Cuenca discovered that discourse markers were found in 17.75% of sentences. The analysis of her corpus compiled from Catalan Historical Review demonstrated that there were six strategies in translating discourse markers including literal translation, omission, addition, generalization, non-literal translation and specification of meaning. Similarly, the process of translating discourse markers from source into target language demonstrates how discourse markers are used by speakers and writers in discourse production. However, the lack of reliance on electronic tools to process the data does not meet strict definitions for corpora in Corpus Linguistics, so the author could have refrained from referring to this study as “corpus-based”.

Discourse markers are inherently polyfunctional due to the many functions they can have (Ghane & Vasheghani Farahani, 2022). In this regard, the tenth chapter which is a study by Ariel, delves into the polyfunctionality of the discourse marker harey in Hebrew language. As a deictic, the discourse marker harey has two main uses: as a cataphoric and as an anaphoric marker to refer to new and old information, respectively. Analysing the written heTenTen and the spoken Haifa corpus, and combining syntactic, semantic and pragmatic features of the discourse marker at stake, Ariel unearths six distinct functions, three for cataphoric and three for anaphoric uses.

There is no denying that there are many positive points this ample ground work of scholarship privileges from. First, it covers, thematically, a wide range of topics in relation to the concept of discourse markers running the gamut of Contrastive Studies, Cognitive Linguistics and Translation Studies. This book is the first tremendous effort which convenes a series of first-hand solid research in relation to Cognitive Linguistics and discourse markers. A scan of the literature will prove that the current book is the first attempt to forge a link between discourses markers and cognitive linguistics. Similarly, this volume must be praised because it investigates a wide range of languages other than English; therefore, its readership is likely to be broad. In addition, this book privileges from thematic unity despite the fact that it fleshes out the concept of discourse markers from various perspectives and in various languages. However, this book is of minimal value, if any, for newcomers to the field. Indeed, the current volume is a good fit for the serious minded, seasoned and advanced researchers and readers in this line of study.