The book is divided into 4 parts, for a total of 10 chapters. It also includes a “glossary of specialist terms for research in quotation”, an author index and an index of terms.
The preface, by the two editors, offers a good review and a basic typology of forms across many languages that have recently been identified as ‘new quotatives’. It identifies a number of recurrent source constructions for these quotatives: comparatives, demonstrative deictics, quantifiers, generic verbs of motion and action. It then shows how extensively this classification overlaps with others as found in the typological literature (e.g. Güldemann 2008). Finally, the editors state their intention of offering a cross-disciplinary perspective, though one that takes in only studies based on attested data. Interestingly, though the title of the volume emphasises the FORMS that introduce quotations, the editors’ essential concern appears to be with quotation itself, their hope being that the collection will “lead to a more holistic perspective on quotation in all its aspects” (p. xxi).
Part I, DISCOURSE PERSPECTIVES, opens with A. Golato’s “Impersonal quotation and hypothetical discourse”, a study of German everyday conversations framed within the methodology of Conversation Analysis. Golato identifies three main functions for the reporting of fictitious discourse, i.e. speech that has not actually been uttered by anyone. It can be used, first, for ‘modelling’ purposes, i.e. to show “what one should or could have said in a given situation” (p. 9), namely for offering advice to a participant reporting a problematic situation; second, to provide support for a claim just made by the speaker; third, to tell humorous stories, often co-constructed by several participants in the conversation. It is noticeable that this opening paper has little to say about quotatives as such, being entirely concerned with the discourse functions of a particular type of speech reports. Rather remarkably, many quotations in Golato’s corpus have no quotative marker at all.
The second paper, I.K. Hasund, T. Opsahl & J. Svennevig’s “By three means. The pragmatic functions of three Norwegian quotatives”, is a spoken corpus-based investigation of the quotative use of ‘bare’ (just), ‘liksom’ (like), ‘sånn’ (such/like (this/that)). As quotatives, they are assumed to fulfil three functions: (i) discourse-organisational (indicating a shift to reported speech), (ii) attitudinal, (iii) interpersonal. The authors show how the quotative functions of the particles take root in their discourse uses (themselves related to their nondiscourse functions), with ‘bare’ typically indicating “heightened speaker involvement” (p. 46) at a key moment in a narrative; ‘liksom’ either serving as an epistemic hedge or inviting the hearer “to visualize or imagine the dramatized situation” (p. 53); and ‘sånn’ expressing “a high degree of epistemic commitment to the exact rendering of the quote” (p. 56). The authors indicate that the most frequent quotative, ‘bare’, may have already been partly neutralised as a general-purpose quotative marker (as may have happened with ‘be like’ in many varieties of English, cf. S. Fox’s contribution). They also show how use of the quotatives differs across generations, providing data demonstrating their ‘novel’ status.
Part II, TYPOLOGICAL PERSPECTIVES, comprises two papers. The first, S. Spronck’s “Minds divided. Speaker attitudes in quotatives” is mainly devoted to the issue of how different languages use quotative constructions to simultaneously report a message and evaluate it. In a detailed and thoughtful paper, the author presents a pilot study (on 21 typologically different languages) of how quotatives should be investigated from a typological perspective. Sentences containing quotatives are understood as quotative constructions, potentially made up of the following lower-level constituents: source construction, SAY construction, Link, and Reportative message. Each is shown to be the potential locus for the expression of (various kinds of) speaker attitudes.
The next chapter, T. Güldemann’s “Thetic speaker-instantiating quotative indexes as a cross-linguistic type”, starts from a typology of quotatives devised in previous work (Güldemann 2008), based on both syntactic (non-clausal-monoclausal-biclausal) and semantic criteria (quote-orientation vs. participant-orientation). Güldemann finds that one type, a non-clausal speaker-oriented quotative index, based on an identificational or presentational construction (± ‘this/here is SPEAKER’), is very frequent across languages, yet has attracted very little attention. This finding invalidates an assumption often made in the past that quotatives usually originate in a construction that has ‘utterance semantics’. Güldemann provides both historical and contemporary evidence of the pervasiveness of this construction and identifies common semantic and formal features. As these quotative indexes foreground the speaker, the speaker nominal does not function as a topic; hence, the construction has ‘thetic’, rather than ‘categorical’ structure. The index normally introduces only direct discourse, and seems to do this in cases which endow the quotation with pragmatic salience and heightened expressivity.
Part III, FUNCTIONAL AND FORMAL PERSPECTIVES, starts with D.Y. Oshima & Shin-ichiro Sano’s “On the characteristics of Japanese reported discourse. A study with special reference to elliptic quotation”. The authors state that proper understanding of ‘quotative predicate ellipsis’ (QPE) allows them to solve two puzzles in Japanese grammar, (i) “why a quotative phrase may co-occur with a nominal direct object under the same predicate”, and (ii) “why a quotative phrase may occur under a predicate that is not a predicate of communication or attitude” (p. 146). They distinguish between five types of QPE, all of them involving ellipsis of a particular form of the verb ‘iu’ ‘say’. Note that in all those cases, the quotative marker ‘to’ (or its informal counterpart ‘tte’) is present after the (directly or indirectly) quoted string. In passing, the authors also argue against the widespread assumption that there exist intermediate forms of quotation in Japanese and that, as a consequence, the distinction between direct and indirect discourse cannot be clear-cut.
L. Vandelanotte’s “Quotative ‘go’ and ‘be like’. Grammar and grammaticalization”, examines how ‘go’ and ‘be like’ came to be used in quotative constructions. He rejects previous accounts for focusing on individual lexemes and argues instead that the locus of change was (incomplete) clauses like ‘X go’ or ‘X be like’. His ‘interclausal’ analysis rests on additional assumptions: (i) that direct quotations are not direct objects of reporting verbs; (ii) that quoting clauses are intransitive. The author argues that the reporting clause is the syntactic head of the construction, though it is conceptually dependent on the quotation. Diachronically, the origin of the quotative is situated in imitative uses of the ‘be like’ and ‘go’ clauses, later analogically extended to the reporting of speech and thought. This process should not be explained in terms of grammaticalisation but rather of ‘constructionalisation’. The author further discusses how ‘be like’ and ‘go’ fit into a wider network of constructions, and considers variants such as ‘be all like’ or ‘go kinda’.
The last paper in Part III is A. Herrmann & M. Steinbach’s “Quotation in sign languages. A visible context shift”. In their German Sign Language corpus, the authors identify four components of the ‘role shift’ that changes the context for the interpretation of (most) indexicals. The first three, in increasing order of frequency, are body lean, head position and eye gaze. Together, they make role shift a nonmanual agreement operator, sharing important properties with the way ‘argument’ verbs agree with their subject and object. In sign languages, verb agreement is realised manually by indicating positions in signing space which encode potential arguments. Role shift, in contrast, marks agreement nonmanually; moreover, it does so not at the propositional level but at the level of the roles ‘(reported) signer’ (marked by the sideward movement part of body lean) and ‘(reported) addressee’ (marked by eye gaze, head position, and the body orientation part of body lean). What of the most frequent signal of quotation, a change in facial expression associated with the reported signer? Since it is not linked to positions in signing space, the authors prefer to leave it outside of role shift proper. It may be that the crucial distinction is between the indexical nature of the three components of role shift as opposed to the iconicity of facial expressions.
Part IV, LANGUAGE VARIATION AND CHANGE, contains two chapters. Sue Fox’s “Performed narrative. The pragmatic function of ‘this is + speaker’ and other quotatives in London adolescent speech” is a variationist and discourse-analytic investigation of the use of quotatives, with special emphasis on ‘be like’ and the brand new ‘this is + speaker’. In her corpus of speech from adolescents and elderly speakers from a culturally and socially diverse area of London, Fox finds that the latter use quotative ‘say’ twice as often as the former. In contrast, ‘be like’ is absent from elderly speech, whereas it is as frequent as ‘say’ in adolescent speech. A multivariate analysis of the data for ‘be like’ reveals no strong associations with sex, with grammatical person, with particular kinds of content (though quotations of non-lexicalised sounds and gestures rank high), or with a particular tense or aspect. At this stage, the innovative quotative ‘this is + speaker’ is functionally much more specialised than ‘be like’, as it mainly serves to mark a dramatic peak in a personal narrative. This specialisation is in keeping with studies of the early uses of other quotatives in various languages. The complementary path of development seems to be exemplified by ‘be like’, which, as it has spread, has acquired a much broader range of functions.
The last chapter, “Dutch quotative ‘van’: Past and present”, by P.-A. Coppen & A. Foolen, shows how the quotative use of ‘van’ evolved from prior uses of this versatile preposition. Quotative ‘van’ is often seen as similar to English ‘be like’ and German ‘zo’. However, ‘van’ is different at least inasmuch as it can also introduce indirect speech reports. Moreover, the three constructions in which ‘van’ introduces direct speech -- ‘van’ triggered by a ‘semantically rich’ verb or noun; triggered by a light verb (+ possibly the adverb ‘zo’); ‘untriggered’ (though often combined with ‘zo’) -- are also found to introduce indirect discourse. Another noteworthy point is the fact that traces of quotative ‘van’ are found as early as the 17th century: some ‘innovative’ quotatives may not be so new after all. What is a fact, however, is that their frequency of occurrence did increase significantly in the last few decades of the 20th century. But here again, there’s something special about ‘van’: it is frequent not just in the language of the youth but also in that of speakers of other generations.
This volume achieves its purpose of providing both a cross-linguistic and a cross-disciplinary perspective on quotatives/quotation. First, the reader connects with the diversity of quotatives across languages, while being shown that quotatives originate in a limited range of markers or constructions. More strikingly perhaps -- and this is what makes this collection different from e.g. Güldemann & von Roncador (2002) -- the reader is confronted with a broad range of approaches to quotatives (and quotation), reflecting distinct research goals, and can experience first-hand why quotatives are of interest to typologists, cognitive and functional linguists, conversation analysts, formal syntacticians, and sociolinguists.
In spite of the broad coverage of the book, two aspects are conspicuous by their absence. First, no link is established between (direct) speech reports and ‘metalinguistic citation’, as in “‘Brussels’ rhymes with ‘muscles’”. Second, and this is probably related, there is hardly any mention of the literature on quotation in the philosophy of language and in formal semantics. In a sense, this is not surprising, as the latter literature is seldom empirically oriented at all (among recent exceptions, let me mention Maier 2012), and itself pays little attention to the various brands of research on display in the present collection.
This not quite unusual situation, in the language sciences, of mutual ignorance is detrimental to progress in the study of quotation. I will give two illustrations: formal semanticists and philosophers, especially the latter, have been at pains to devise a typology of quotation for more than a decade (certainly since Cappelen & Lepore 1997 and Recanati 2001), and they have convincingly shown that categories like ‘mixed quotation’ or ‘open quotation’ need to be recognised as distinct from (and maybe intermediate between) direct speech and indirect speech. It is striking that these categories are alluded to by very few of the contributors to this volume. Vandelanotte mentions one example, and Oshima & Sano reject the analysis that Japanese has a form of speech report intermediate between direct and indirect discourse. When Golato (p. 20) writes that “gestures and bodily postures associated with the talk tend not to appear in indirect discourse”, there is a possible suspicion that this conclusion may result from neglect of hybrid forms.
On the other hand, though ‘hybrid’ forms of quotation have received quite some attention in the philosophy of language and formal semantics (e.g. De Brabanter 2005, 2010), scholars have generally limited themselves to a two-term opposition between mixed and scare quoting, failing to recognise the sheer variety of possibilities that in fact exist (cf. Evans 2013). The studies in the present collection provide further evidence of the diversity of quotation, and of the many forms that can introduce them, and should be taken seriously by any theorist attempting a classification of quotation types.
When reading a collection of articles, one may find reasons to disagree with (at least) details in most of the contributions, and this is not the place for a complete overview of questions or objections. I will limit myself to a few comments on a paper that I regard as a strong one in the volume, because they again illustrate why even more integration of distinct fields of research may prove useful. Vandelanotte makes a good case in favour of an interclausal analysis of quotative constructions involving direct speech. However, it is not clear that he is right to provide the same account for reporting clauses (i) in initial position and (ii) in medial or final position. Using a classification that cuts across the ‘Direct/Indirect’ opposition, Recanati (2001) classifies the first cases as ‘closed’ quotations (filling a syntactic slot in the matrix clause) and the others as ‘open’ (where the quotation is syntactically independent). It would be too long to substantiate that distinction, but consider the following data, suggesting the reasonableness of differentiated treatment: “She said ‘I don’t know’, didn’t she?” vs. *“‘I don’t know’, she said, didn’t she?”. There is also the fact that “She said this” is nothing unusual, whereas “This she said” is unusual (at least, it is much more marked than “‘I don’t know’, she said”). Consider, too, the difference in acceptability between “She said ‘I don’t know’” and “‘I don’t know’, she said”, as answers to “What did she say?”.
Let me close with a brief word on the readership: none of the papers in the collection is very technical in content and format. Taken individually, (at least some of) the papers are already accessible to undergraduate students in linguistics. However, given the many traditions and methodologies brought together here, only a seasoned linguist -- graduate level at least -- will be able to derive a major benefit from the collection as a whole. Even there, it will take some open-mindedness and curiosity to go off the beaten track and take in all the papers. But that, after all, is certainly not an illegitimate demand to make. And there is no doubt that reading this volume will then prove a very rewarding experience.
Cappelen, Herman, and Ernie Lepore. 1997. Varieties of Quotation. Mind 106.429-50.
De Brabanter, Philippe (ed.). 2005. Hybrid Quotations. Amsterdam: John Benjamins. (Belgian Journal of Linguistics 17.)
De Brabanter, Philippe. 2010. The semantics and pragmatics of hybrid quotations. Language and Linguistics Compass 4-2.107-120.
Evans, Nicholas. 2013. Some problems in the typology of quotation: a canonical approach. In D. Brown, M. Chumakina & G. G. Corbett (eds.), Canonical Morphology and Syntax. Oxford: O.U.P, 66-98.
Güldemann, Tom. 2008. Quotative indexes in African languages: a synchronic and diachronic survey. Berlin: Mouton de Gruyter. (Empirical Approaches to Language Typology 34.)
Güldemann, Tom, and Manfred von Roncador (eds.). 2002. Reported discourse: a meeting ground for different linguistic domains. Amsterdam: Benjamins. (Typological Studies in Language 52.)
Maier, Emar. 2012. Switches between direct and indirect speech in Ancient Greek. Journal of Greek Linguistics 12.118-139.
Recanati, François. 2001. Open quotation. Mind 110.637-87.
| ABOUT THE REVIEWER:
Philippe De Brabanter teaches English linguistics at the Université Libre de Bruxelles. He is an associate member of Institut Jean Nicod (Paris), where he was a postdoctoral fellow. He wrote his PhD on the semantics and pragmatics of everyday metalanguage (2002). He has published on quotation (e.g. “Hybrid Quotations”, Belgian Journal of Linguistics, 17, 2005 (ed.)), on metalinguistic anaphora and gestures mixed with words. His current research interests are quotation theory, the depictive use of language, various issues at the interface between semantics and pragmatics, and colour adjectives and nouns.