LINGUIST List 29.3638

Fri Sep 21 2018

Review: General Linguistics; Linguistic Theories; Syntax: Hengeveld, Narrog, Olbertz (2017)

Editor for this issue: Jeremy Coburn <>

***************** LINGUIST List Support *****************

Fund Drive 2018
28 years of LINGUIST List!
Please support the LL editors and operation with a donation at:

Date: 12-Apr-2018
From: Brendon Yoder <>
Subject: The Grammaticalization of Tense, Aspect, Modality and Evidentiality
E-mail this message to a friend

Discuss this message

Book announced at

EDITOR: Kees Hengeveld
EDITOR: Heik Narrog
EDITOR: Hella Olbertz
TITLE: The Grammaticalization of Tense, Aspect, Modality and Evidentiality
SUBTITLE: A Functional Perspective
SERIES TITLE: Trends in Linguistics. Studies and Monographs [TiLSM]
PUBLISHER: De Gruyter Mouton
YEAR: 2017

REVIEWER: Brendon Yoder, University of California, Santa Barbara


This edited volume includes ten papers discussing the grammaticalization of tense, aspect, modality, and evidentiality (TAME) from a functional perspective. After a brief introduction by the editors, the first three main papers are general discussions of the grammaticalization of TAME, while the remaining seven chapters are language-specific studies. The theory of Functional Discourse Grammar (Hengeveld & Mackenzie 2008) features prominently in the volume. This framework as applied to the grammaticalization of TAME is developed extensively in the chapters by Hengeveld and Giomi and is discussed at least briefly in all the remaining chapters.

In his chapter “A hierarchical approach to grammaticalization” (pp 13-38), Kees Hengeveld argues for Functional Discourse Grammar (FDG) as a framework for modeling grammaticalization. The author gives a brief overview of FDG and then shows how grammaticalization fits with the model, using examples of grammaticalization of TAME throughout. The FDG model of grammar is hierarchical, with a pragmatic (“interpersonal”) level and semantic (“representational”) level. Within each level are nested layers, with each successive layer having wider semantic scope than the layer within it. Hengeveld argues that grammaticalization always involves a scope increase in function; that is, contentive change always results in wider scope. Examples of layer changes at the representational level are given, while at the interpersonal level a single possible English example of layer changes is given. Functional scope can also increase through a change from the representational level to the interpersonal level, but not vice versa. Hengeveld also argues that form and function do not proceed in tandem in the grammaticalization process, although the two are related in a relative way. At any given stage the function might grammaticalize further (increase in scope), while the form might also grammaticalize further (decrease in lexicality and formal independence) or remain the same. Conversely, at any given stage the form might grammaticalize further, while the function might also grammaticalize further or remain the same. The crucial point is that there is no backward movement along the cline. Discussion of TAME is limited to examples showing how FDG can model grammaticalization.

The chapter by Riccardo Giomi (pp 39-74) titled “The interaction of components in a Functional Discourse Grammar account of grammaticalization” follows up on Hengeveld’s chapter by presenting a more in-depth look at an FDG account of grammaticalization. While Hengeveld discusses primarily the Grammatical Component of FDG, with its two levels and various layers, Giomi proposes an integration of the other two components of FDG into an FDG account of grammaticalization. These two are the Conceptual Component and the Contextual Component. Of particular importance is the Contextual Component, where grammar-external information relevant to the grammatical categories of the language is modeled. Grammaticalization is seen as a shift from the Contextual Component to the Grammatical Component in a process of conventionalization. Conventionalization is “the process whereby a context-specific inferential meaning is reanalyzed as a new grammatically encoded function” (47). The bulk of the chapter is a presentation of this reanalysis process, largely following Heine (2002). In this earlier work four stages are identified: initial stage, bridging context, switch context, and conventionalization. The initial stage and conventionalization stage represent the start and end points of the grammaticalization process, respectively. The bridging context is one in which a new meaning is possible through pragmatic inference in a particular context. In the switch context, the new meaning is now possible in a context incompatible with the source meaning. Heine’s four stages are followed exactly here, simply being cast in an FDG framework. At the end of the chapter two case studies are presented. The first is the grammaticalization of Old English <sculan> ‘owe’ into the future temporal marker <shall>. The second is the change in Modern Greek <tha> from future tense to subjective epistemic modality. Like the previous chapter, this chapter’s primary focus is the adequacy of FDG for modeling grammaticalization, with illustrations of the grammaticalization of TAME.

Heiko Narrog’s chapter “Relationship of form and function in grammaticalization – the case of modality” (pp 75-110) discusses grammaticalization of both form and function, and the relationship between the two. Like the previous two authors, he argues that grammaticalization of function is primary while grammaticalization of form is “rather epiphenomenal” (75). Also like the previous two authors, he argues that grammaticalization of form and function need not proceed in tandem, but in most cases an advancement in one means that the other will either advance as well or remain the same. Grammaticalization can begin at any point on a hierarchy of functional categories; because of this, a correlation between form and function in grammaticalization only makes sense for individual constructions. The author cites his previous work where this has been argued for at length (Narrog 2005; 2012), but he presents it again here because of a persistent assumption in grammaticalization studies that cross-linguistic correlations of form and function can be established. Grammaticalization of the form and function of an individual construction in a language is expected to proceed unidirectionally, but since the starting point for each construction is different, comparison across constructions is not productive. Formal grammaticalization is seen as the result of online, rather mechanical processes of phonological and morphological reduction. The cognitive and social mechanisms of the grammaticalization of function, that is, the shift from pragmatic inference in a specific context to grammatically encoded function, remain to be explored. Narrog offers a few possible lines of exploration into these mechanisms. These include Haspelmath’s (1999) notion of speaker “extravagance” and the possibility that extension of context is cognitively perceptible while reduction is not. He also suggests that social factors in language change might be relevant. Examples of grammaticalization of modal constructions are presented from English, Chinese, Japanese, and Korean.

The chapter by Sophie Villerius (pp 111-132), “Modality and aspect marking in Suriname Javanese: Grammaticalization and contact-induced change”, is the first language-specific study in the volume. The author discusses contact-induced and language-internal changes in the modal and aspectual auxiliaries of Suriname Javanese. Javanese, an Austronesian language spoken primarily in Indonesia, was brought to Suriname by Javanese laborers around the beginning of the 20th century, where it is now in intensive contact with the other languages of the country, primarily Dutch and Sranan Tongo. Based on a comparison of small corpora of Indonesian Javanese and Suriname Javanese, this paper examines two changes in modal markers and two changes in aspect markers. The changes in modal markers appear to be strictly formal, with no functional change. The first is a lexical borrowing of the Sranan modal auxiliary <proberi> ‘try’, which replaces the Javanese modal auxiliary <jajal>. The second is a language-internal change, the replacement of the auxiliary marking ability <isa> with another auxiliary <inter>. The first aspect change involves a change in function of the auxiliary <arep>, away from future in Indonesian Javanese toward prospective aspect ‘be going to’ in Suriname Javanese. The final change, which is still tentative given the paucity of data, is a possible formal shift in the marking of progressive aspect from the auxiliary <lagi> to a type of presentational structure with an existential verb. This paper is informative in that it brings together studies of contact-induced change, primarily Matras (2007) and Sakel (2007), with studies of (language-internal) grammaticalization, primarily Hengeveld (2011). The research in this chapter could be enhanced by increasing the size and diversity of the corpora of both Indonesian Javanese and Suriname Javanese. It might also benefit from comparing Suriname Javanese with the form of Javanese that was spoken by the migrant workers who first made their way from Indonesia to Suriname rather than with modern Indonesian Javanese, which has been in intense contact with Indonesian (Errington 1998) and thus may not be the best baseline for comparison unless this contact situation is also considered.

The chapter by Lotta Jalava, “Grammaticalization of modality and evidentiality in Tundra Nenets” (pp 133-161), proposes two diachronic pathways for the development of modal and evidential suffixes in Tundra Nenets. The first primary pathway is the finitization of participial predicates, resulting in most of the modal and evidential suffixes in the language. The second pathway is insubordination (Evans 2007), in which a subordinate clause comes to be used as a main clause. These two diachronic pathways correspond nicely to the synchronic morphosyntax. The modal and evidential suffixes resulting from finitization of participial predicates are part of a single formal class of verbal suffixes, all occurring in a single suffix slot. The single evidential suffix resulting from insubordination has quite different structural properties, resembling its diachronic predecessor. Verbs with this evidential are not marked for tense, and the person suffixes on these verbs are identical with possessive pronominal forms. The various modal and evidential suffixes are of varying ages. Some are recent developments in Tundra Nenets, with both the older participial and new finite verb forms attested in the modern language. For others, only the new form is attested. Diachronic pathways are postulated based on cognates in other Samoyedic languages and previous reconstructions of Proto Samoyedic. This chapter contributes to Uralic studies in being the first to propose a pathway for the development of the Tundra Nenets modal and evidential suffixes. It also contributes to the broader typological theme of this volume in providing examples of grammaticalization of both the form and function of TAME markers. Confirming the position of the first three authors in this volume (Hengeveld, Giomi, Narrog), both the form and the function of the Tundra Nenets suffixes developed unidirectionally, but not necessarily in tandem.

The chapter by Shadi Davari and Mehrdad Naghzguy-Kohan (pp 163-189), titled “The grammaticalization of progressive aspect in Persian”, traces the development of the Persian lexical verb <dâštan> ‘have’ into a progressive auxiliary. Cross-linguistically, progressive constructions commonly develop from lexical items expressing location and motion, while development from a possessive construction as in Persian is unusual. The authors demonstrate clearly how this grammaticalization happened, showing how both the form and the function of possessive <dâštan> developed to a point where reanalysis as a progressive was just a small step. In terms of form, the point of crossover from possessive to progressive occurred in relative clauses that shared a subject with the matrix clause. In terms of function, <dâštan> developed from indicating temporary possession, or ‘being with’ a concrete entity, to indicating ‘being with’ a process or event. After reanalysis, the construction spread beyond same-subject relative clauses to other parts of the grammar. At the end of the paper the authors present their Auxiliation Dimensions Model, the details of which are being written up in another publication (Davari and Naghzguy-Kohan, forthcoming). This model considers three dimensions of auxiliation: the force (functional motivation), the lexical source of the auxiliary, and the degree of auxiliation from full lexical verb to highly grammaticalized auxiliary. It is not entirely clear from the short presentation here how the Auxiliation Dimensions Model reflects processes specific to the grammaticalization of auxiliaries as opposed to grammaticalization pathways in general.

In his chapter “Grammaticalization as morphosyntax and representation: Mood from tense markers in the Old Irish and Romance conditional” (pp 191-214), Carlos García Castillero discusses the development of the conditional construction in Western Romance languages and Old Irish, with comparison to the English conditional construction. In all three languages, a previous temporal meaning of future-in-the-past (e.g. ‘She told me that she WOULD SEE’) came to be reanalyzed with a hypothetical modal meaning in the main clause of a conditional construction (e.g. ‘If she went, she WOULD SEE’). García Castillero notes that this reanalysis involves a change from subordinate clause to main clause, but this is actually a rather small step. The source construction, a complement clause of verbs of speaking and perceiving, cross-linguistically tends to have fewer features of subordinate clause syntax than other types of subordinate clauses and is a frequent source of insubordination (Evans 2007). Further, the resulting construction, while syntactically a main clause in all three languages, is semantically dependent on the adverbial clause containing the condition. In all these languages, the source construction involved a combination of past tense marking with past-imperfective (‘imperfect’) marking, leading the author to suggest that this recurrent form-function pairing in grammaticalization involves a “relatively straightforward diachronic path” (192). While the formal changes were broadly similar, the grammatical details of each language gave different results. In English, the resulting periphrastic construction involves an auxiliary and a main verb. In Western Romance, the conditional is marked with a suffix resulting from the coalescence of the past and imperfect suffixes. In Old Irish, the two previously existing morphemes were combined to serve a new function. García Castillero proposes that this represents a new kind of morphologization that calls for a broader definition of the term. Previous studies discuss morphologization as either coalescence of previous lexical items or as reanalysis of morphophonological alternations as functionally distinct. By including the development of the Old Irish conditional construction, the proposed broader definition of morphologization is “every diachronic process leading to the creation of a morpheme, whether based on independent or dependent lexical elements, phonological features, or on the combination of previously existing morphemes which comes to express a new meaning” (209).

Jimena Tena Dávalos in “The end of a cycle: Grammaticalization of the future tense in Mexican Spanish” (pp 215-239) presents a history of the development of future tense marking in Mexican Spanish, including a corpus study showing currently emerging developments. In the history of the language, future tense marking has gone through a cyclical process where an analytic future construction develops into a synthetic form, the synthetic form develops other functions such as expression of epistemic modality, and a new analytic future construction emerges. Classical Latin had a (synthetic) future suffix. This was replaced by an analytic construction with the auxiliary <habere> ‘have’; this analytic construction, in turn, grammaticalized into verbal inflections in Western Romance languages. Now an analytic construction with ‘go’ + infinitive has emerged alongside the synthetic future construction. The main contribution of this study is an analysis of a diachronic corpus showing the functional development of the synthetic and analytic future constructions over the course of the past 600 years. The synthetic future was originally used almost exclusively with future time reference, while in current Mexican Spanish about 25% of occurrences express modal meanings. The newer analytic form with the verb ‘go’ was originally used to indicate motion. In an intermediate stage it was used most frequently to indicate prospective aspect, while current Mexican Spanish speakers use the form most frequently with simple future reference and secondarily to indicate immediate future or prospective aspect. The author compares these findings with the model of FDG, which predicts that grammaticalization always involves an increase in functional scope. The functional scope of TAME marking as defined in the model fits with the development of the synthetic and analytic future markers in Mexican Spanish.

The chapter by Aude Rebotier, “The grammaticalization of tenses and lexical aspect – the case of German and French perfects” (241-272), traces the development of perfect constructions in French and German as related to lexical aspect. While previous work has discussed the grammaticalization of French and German perfects, the unique contribution of this paper is that it explicitly compares the grammaticalization pathways of the perfect in these two languages. The semantic grammaticalization of the perfect proceeded in broadly the same way in the two languages: An adjectival construction with ‘have’ or ‘be’ developed a resultative function, which in turn developed an anterior usage. In the final stage it developed a temporal function as a past tense marker. In the first stage, the adjectival construction was restricted to ‘transformative’ or change-of-state verbs, a subcategory of Vendler’s (1957) category of accomplishments. As the construction developed resultative and anterior functions, the lexical restrictions loosened and verbs with other lexical aspects began to be used. The process is complete with the past tense usage, which can occur with verbs of all lexical aspects. Formal grammaticalization of this construction in both French and German involved a loss of agreement on the participle: object agreement with ‘have’ constructions and subject agreement with ‘be’ constructions. Another grammaticalization process is semantic bleaching of ‘have’ and ‘be’, where the distinction between transitive and intransitive is lost and becomes lexically specified allomorphy; in some cases, one of the two morphemes is lost. Rebotier concludes that the French perfect is semantically more grammaticalized than the German perfect, but formally less so. However, the development of each construction proceeded in the same direction as predicted in FDG.

The final chapter in the volume, “The grammaticalization of Dutch moeten: modal and post-modal meanings” by Hella Olbertz and Wim Honselaar (pp 273-300), discusses semantic changes in the Dutch modal auxiliary <moeten>. Unlike most other Germanic languages that have multiple modals, this is the only modal in Dutch. This marker is highly polysemous and is used to express various epistemic and deontic modalities. The authors develop a semantic taxonomy of the modal distinctions expressed by <moeten> that combines Narrog’s (2005) volitive vs. non-volitive modality and a modified version of Hengeveld’s (2004) distinction involving participant, event, episode, and proposition. This yields an eight-way distinction that can be mapped onto the layered hierarchical levels of FDG. The authors then trace the semantic development of <moeten> from its earliest documented attestations where it indicated participant-oriented possibility. An additional sense indicating event-oriented possibility developed, followed by a sense indicating deontic necessity. <moeten> also developed the possibility of occurring independently of another verb, with the sense ‘must go’. The authors argue against a proposal by Nuyts (2013) that this represents a case of degrammaticalization, stating that it is better viewed as a case of lexicalization. Another relatively recent development is the use of <moeten> in an immediate imperative with perception verbs. The authors end the chapter by showing how the semantic developments of <moeten> follow the predictions of FDG in that each change involved an increase in scope.


The chapters in this volume bring together the functionally-oriented literature on TAME (Comrie 1976, 1985; Dahl 1985; Palmer 2001; Narrog 2005; inter alia) with the literature on grammaticalization (Heine & Kuteva 2002; Hopper & Traugott 2003; Heine & Narrog 2011; inter alia). The landmark study on grammaticalization of TAME is Bybee, Perkins & Pagliuca (1994), which is referenced frequently throughout the book. The book contains important insights that are sure to move forward our understanding of the grammaticalization of TAME.

Three rather minor deficiencies in this volume present themselves. First, the emphasis on FDG is somewhat unexpected. Given the research context in which this book is situated, as well as the title and introduction that situate the book as a functionally oriented volume on the grammaticalization of TAME, the reader is not prepared for extensive studies of FDG unless, of course, one is already acquainted with the work of individual scholars. The book title, introduction, and summary on the back cover and the de Gruyter website would more accurately reflect the books’ contents if they explicitly mentioned the prominent place of FDG in many of the volume’s chapters.

Second, a less obvious mismatch between the introduction and the main chapters concerns the typological diversity of languages analyzed by the chapter authors. The introduction argues for the importance of languages “of maximally different types and affiliations” (5) in a functional approach, presenting this volume as containing a typologically diverse sample of languages. The language-specific studies are ordered with those most distant from Western European first, and studies of Western European languages last. However, of the nine languages represented in studies of individual languages, seven are Indo-European. Of those, five are Germanic or Romance. The only non-Indo-European languages represented are Javanese (Austronesian) and Tundra Nenets (Uralic). Readers might be better served with an explicit statement that the sampling is skewed toward well-studied languages and a suggestion that further studies include a broader sampling of languages.

Third, a minor formatting addition to aid readers would be division of the table of contents into two sections: section one containing the three general studies and section two containing the language-specific studies. This would make the contents more quickly accessible.

In spite of the rather minor limitations mentioned above, this volume contains careful, high quality research by both leading and emerging scholars in the field and provides an important contribution to the study of the grammaticalization of TAME. It is important reading for scholars working on grammaticalization or on TAME, as well as those interested in the framework of FDG. Studies of individual languages are likely to be of interest to areal and language family specialists, especially those whose research focuses on the historical development of the language(s) in question.


Bybee, Joan L., Revere Perkins & William Pagliuca. 1994. The evolution of grammar: Tense, aspect, and modality in the languages of the world. Chicago: University of Chicago Press.

Comrie, Bernard. 1976. Aspect. Cambridge: Cambridge University Press.

Comrie, Bernard. 1985. Tense. Cambridge: Cambridge University Press.

Davari, Shadi & Mehrdad Naghzguy Kohan. Forthcoming. The emergence of auxiliaries in Persian.

Errington, J. Joseph. 1998. Shifting languages: Interaction and identity in Javanese Indonesia. Cambridge: Cambridge University Press.

Evans, Nicholas. 2007. Insubordination and its uses. In Irina Nikolaeva (ed.), Finiteness, 366–431. Oxford: Oxford University Press.

Haspelmath, Martin. 1999. Why is grammaticalization irreversible? Linguistics 37(6). 1043–1068.

Heine, Bernd. 2002. On the role of context in grammaticalization. In Ilse Wischer & Gabriele Diewald (eds.), New reflections on grammaticalization, 83–101. (Typological Studies in Language 49). Amsterdam: John Benjamins.

Heine, Bernd & Tania Kuteva. 2002. World lexicon of grammaticalization. Cambridge: Cambridge University Press.

Heine, Bernd & Heiko Narrog (eds.). 2011. The Oxford handbook of grammaticalization. Oxford: Oxford University Press.

Hengeveld, Kees. 2004. Illocution, mode and modality. In Geert Booij, Christian Lehmann & Joachim Mugdan (eds.), Morphology: a handbook on inflection and word formation II, 1190-1202. Berlin & New York: Mouton de Gruyter.

Hengeveld, Kees. 2011. The grammaticalization of tense and aspect. In Bernd Heine & Heiko Narrog (eds.), The Oxford handbook of grammaticalization, 580–594. Oxford: Oxford University Press.

Hengeveld, Kees & J. Lachlan Mackenzie. 2008. Functional Discourse Grammar: A typologically-based theory of language structure. Oxford: Oxford University Press.

Hopper, Paul J. & Elizabeth Closs Traugott. 2003. Grammaticalization. 2nd ed. (Cambridge Textbooks in Linguistics). Cambridge: Cambridge University Press.

Matras, Yaron. 2007. The borrowability of structural categories. In Yaron Matras & Jeanette Sakel (eds.), Grammatical borrowing in cross-linguistic perspective, 31–74. Berlin: Mouton de Gruyter.

Narrog, Heiko. 2005. Modality, mood, and change of modal meanings: A new perspective. Cognitive Linguistics 16(4). 677–731.

Narrog, Heiko. 2012. Modality, subjectivity, and semantic change: A cross-linguistic perspective. Oxford: Oxford University Press.

Norde, Muriel. 2009. Degrammaticalization. Oxford: Oxford University Press.

Nuyts, Jan. 2013. De-auxiliarization without de-modalization in the Dutch core modals: A case of collective degrammaticalization? Language Sciences 36. 124–133.

Palmer, F.R. 2001. Mood and modality. 2nd ed. Cambridge: Cambridge University Press.

Sakel, Jeanette. 2007. Types of loan: Matter and pattern. In Yaron Matras & Jeanette Sakel (eds.), Grammatical borrowing in cross-linguistic perspective, 15–30. Berlin: Mouton de Gruyter.

Vendler, Zeno. 1957. Verbs and times. The Philosophical Review 66(2). 143-160.


Brendon Yoder is a member of SIL International and a PhD candidate in linguistics at the University of California Santa Barbara. His research focuses on documentation and description of Abawiri, a Papuan language of Indonesia. Recent projects include collaboration with the community on an orthography for community use, analysis of the tone system, and investigation of grammatical relations and their connection to pragmatic structuring of discourse. He is currently compiling a corpus of video-recorded texts with a grant from the Endangered Language Fund. He has also conducted fieldwork on phonetics and phonology in both Nias and Enggano, Austronesian languages of Western Indonesia.

Page Updated: 21-Sep-2018