Review of Synchrony and Diachrony
The present volume stems from the workshop “Gradualness in change and its relation to synchronic variation and use”, organized by the editors, Anna Giacalone Ramat, Caterina Mauri and Piera Molinelli at the University of Pavia (Italy) in 2011.
As the editors put it in the introduction, the aim of the volume is “to investigate the mutual relations between synchrony and diachrony, in order to shed light on their interface” (p. 1). One of their main claims is that certain methodological tools, among them semantic maps and constructional approaches, can be applied both to diachronic and synchronic phenomena, which helps us understand the relationship between the two dimensions. According to the editors, the label ‘synchrony-diachrony interface’ can be applied to all those cases where a phenomenon cannot be accounted for without taking into consideration both diachronic and synchronic variation. Thus, it involves both the linguistic data itself and the linguist’s perspective on this data. One of its most widely recognized instances is the relationship (however controversial it may be) between gradience and gradualness as features of synchronic variation and diachronic change, respectively. In fact, a lot has been done within historical linguistics, dialectology, grammaticalization, language variation and change and other fields to assess the diachronic implications of synchronic variation. Analogy should be also considered a manifestation of the synchrony-diachrony interface. But little has been done yet to provide a unified analysis of such phenomena, with an appropriate focus on methodological and theoretical issues. The volume under review is intended to fill this gap. The papers adopt different frameworks, including Construction Grammar and Generative Grammar.
All of the languages covered are European. Most papers draw data from Germanic: English is considered by Trousdale, Margerie, Disney, van der Auwera, van de Pol & Cuyckens; German data is analysed by van der Auwera; Dutch, by De Vos and Semplicini; Swedish, by Rosenkvist & Skärlund. Romance is represented by Latin (Magni; Fedriani, Manzelli & Ramat), Ladin (Wratil), Italian (Voghera), and South Calabrian dialects (De Angelis). Two papers are dedicated to other Indo-European languages, Modern Greek dialects and Welsh, by Melissaropoulou and Currie, respectively. Only one paper, by Egedi, treats a non-IE language, Hungarian. Finally, two articles discuss linguistic areas: the Circum-Mediterranean area is in the focus of Fedriani, Manzelli & Ramat, whereas Wratil discusses the Alpine languages, comprising a number of Romance and Germanic varieties.
The book is divided in three parts, each of which is centered on some particular problem pertaining to the synchrony-diachrony interface.
The first part, “The role of analogy and constructions in the synchrony-diachrony interface”, discusses cases of diachronic change driven by synchronically available elements of variation. It opens with “Gradualness in language change: A constructional perspective”, by Graeme Trousdale, who aims to reconcile the hypotheses of gradualness and abrupt reanalysis. The author argues that it is inaccurate to see gradualness as the diachronic equivalent of synchronic gradience. It is demonstrated that the perception of gradualness can depend on the fact that the process of constructionalization might consist of a series of micro-steps, each, in turn, a case of abrupt neoanalysis. The author presents a case study, consisting of a constructional, corpus-driven account of the development of the English preposition ‘during’ from the Middle English present participle ‘duren’, ‘during’, derived in turn from French ‘durer’ (from Latin ‘durare’, ‘to be hard, hold out, last’).
In “Gradual change and continual variation: The history of a verb-initial construction in Welsh”, by Oliver Currie, two different analyses of the development of Absolute-initial verb (AIV) order in Early Modern Welsh are compared, a diachronic Construction Grammar account and a Principles and Parameters analysis. The AIV order was a marginal construction in Middle Welsh, but appears to be the dominant word order in several Early Modern Welsh texts, and its definitive establishment was preceded by a protracted period of variation. The abovementioned theoretical perspectives differ greatly with respect to the question of the nature of syntactic change (gradualness vs. discreteness) and the problem of syntactic variation. The Principles and Parameters framework envisages syntactic change as discrete, whereas historical data suggests gradual patterns. The author argues that a Construction Grammar account of the development of the AIV order in Welsh as a gradual increase in the frequency of use of the construction under analysis fits the data better. The process was motivated by sociolinguistic and stylistic factors and did not involve either grammaticalization or change in the meaning or function of the construction. Given this, Traugott and Trousdale’s (2008) model of gradualness as a sequence of discrete and therefore abrupt micro-steps at the level of individual speakers, resulting in gradual change at the level of community, is not relevant here.
In “Can you literally be scared sick? The role of analogy in the rise of a network of Resultative and Degree Modifier constructions”, by Hélène Margerie, the micro-construction ‘NP1 scare NP2 sick / NP be scared sick’ is examined in the light of comparable diachronic changes, with the use of corpus and internet data. It appears that the micro-construction in question does not follow the pathway of other resultative constructions of the form ‘NP1 VB NP2 / NP be ADJ XP’, which were historically reanalyzed into Degree Modifier constructions. The author concludes that the resultative meaning of the ‘NP1 scare NP2 sick / NP be scared sick’ construction is an outcome of analogical change, shaped on the model of formally and functionally related constructions.
“The reputed sense of ‘be meant to’: A case of gradual change by analogy”, by Steve Disney, is a case study in a usage-based construction grammar perspective. In its evidential ‘reputed’ sense, it is not a passive counterpart of ‘mean’. The paper discusses the development of this evidential use by analogy with ‘hearsay’ NCI (nominativus cum infinitivo) constructions. ‘Be meant to’ has such acknowledged meanings as intention, weak obligation and predestination/design. Moreover, a novel evidential ‘reputed’ sense has recently evolved in British English, together with ‘be supposed to’, the latter having multiple synonymy with the former. Now, the construction ‘be meant to’ can express a reported belief about an expected future. According to the author, the developmental path of ‘be meant to’ challenges the Semantic Map Connectivity Hypothesis (Croft 2001), in that the construction under analysis seems to ‘miss’ a sense in the course of its development.
In “Gradualness in analogical change as a complexification stage in a language simplification process: A case study from Modern Greek dialects”, by Dimitra Melissaropoulou, cross-paradigmatic levelling in the nominal system is analysed as a gradual process, leading to grammar simplification. Several Greek dialects are discussed, namely Aivaliot, Lesvian, Pontic, Livisi, and Silli. The focus falls on the intra-dialectal role of markedness, allomorphy, and case syncretism. There appears to be a tendency for the loss of inflection class sub-paradigms and the establishment of uniform inflectional patterns, i.e. greater paradigmatic simplicity. The author aims to show that the direction of change can be predicted on the basis of synchronic variation and, moreover, intra-dialectal variation can represent different stages of the same change.
The second part, “Synchronic variation and language change”, focuses on synchronic variation as the source and result of diachronic change; in other words, it discusses synchronic variation as motivated by or, on the contrary, motivating language change.
In “Semantic maps, for synchronic and diachronic typology”, by Johan van der Auwera, the advantages of classical, or ‘connectivity’ semantic maps are discussed compared to more recent statistical maps, and the possibilities offered by the latter for studying diachronic processes are illustrated. The ‘old’ type of semantic maps has the feature of linking, which is crucial for the representation of connections between uses. As for statistical maps, they show only the proximity of the contexts, not the connections. What is more, connectivity maps can represent theories, as they are predictive and falsifiable, whereas proximity maps can be used only for generalizations concerning certain contexts in which certain constructions occur. Van der Auwera defends the connectivity approach and claims that, although both types of semantic maps have merits, a major strong point of connectivity maps is that they offer an insightful model of synchronic variation and at the same time promote a diachronic perspective.
“Synchronic gradience and language change in Latin genitive constructions”, by Elisabetta Magni, is dedicated to Latin adnominal constructions. It is well-known that there is no strict borderline between possession in a broad sense (anchoring relations) and other semantic types of adnominal modification (non-anchoring relations). In Latin, with its flexible word order, genitive constructions exhibit both G(enitive)N(noun) and NG patterns. As for their semantics, they can convey possessive meanings, when the possessor performs the role of a pragmatic anchor (or, in other words, a reference point) for identifying the possessee, but they can also express non-anchoring relations, being potentially ambiguous between different readings. It competes with another widespread adnominal construction, namely, ‘Noun + denominal adjective’, and their functions converge to a certain extent. After a long period of constructional variation, NG structures gradually oust constructions with denominal adjectives and become the preferred means for expressing non-anchoring relations.
“Double agreement in the Alpine languages”, by Melani Wratil, provides an account of double agreement effects in several linguistic varieties: Bavarian, Alemannic and Ladin. Alpine languages exhibit agreement allomorphy in verbal inflection, reflecting an intermediate stage in the grammaticalization of atonic subject pronouns to verbal agreement suffixes. Analyzing the phenomenon in a generative perspective, the author claims that its emergence gave rise to more economical structures, as the highly specialized pronominal and verbal paradigms brought about the least costly syntactic derivations and the least redundant representations which can be compatible with the Primary Linguistic Data.
In “On variation in gender agreement: The neutralization of pronominal gender in Dutch”, by Lien De Vos, the triggering role of competition between syntactic and semantic factors for variation in gender agreement is accounted for. The northern Dutch category of gender is known to have evolved from a three-term system to a two-term one, as masculine and feminine fused into one gender. As for Southern Dutch, although it retains the original three-gender system, it appears to be in transition from a grammatical gender system towards a more semantic one. Evidence from a corpus of spoken Dutch, ‘Corpus Gesproken Nederlands’, is provided to support the claim that in Southern Dutch the antecedent’s position in the Givenness Hierarchy and its syntactic role influence the use of gender-marked pronouns. The synchronic variety is explained diachronically as depending on increased structural ambiguity, caused by the reduction of gender-marking morphology.
In ‘Synchronic variation and grammatical change: The case of Dutch double gender nouns’, by Chiara Semplicini, one more aspect of gender in Dutch is investigated: double gender nouns (DGNs), a category which, on the one hand, seems to be a marginal phenomenon, but, on the other hand, appears to be persistent in the history of the language. DGNs tend to form kinds of semantic clusters and include such referents as objects and substances, as well as abstract terms. The number of nouns involved constantly increased from the Middle Ages, when the overt gender markers and, consequently, transparency of the inflectional system were lost, up to the modern period, but decreased drastically in the 20th century due to the process of systematization and standardization. The author provides additional evidence of the gradual loss of grammatical gender in Dutch in favor of a semantic system and to explain the diachronic persistence of the phenomenon under study.
In ‘A case study on the relationship between grammatical change and synchronic variation: The emergence of tipo[-N] in Italian’, by Miriam Voghera, a diversity of non-nominal uses of Italian ‘tipo’ (‘type’), exhibiting a complex network of syntactic and pragmatic functions, is classified and analyzed in terms of diachronic development and synchronic variation. It is shown how the noun underwent the decategorization process, first developing the function of adnominal modifier in the first half of the 20th century and acquiring, later on, novel uses, including as similative marker, approximator, interclausal connector and focus marker. The most characteristic features of the phenomenon is the retention, up to the present, of uses which emerged at different stages (although frequency and relevance of each of them have changed significantly over time) and fuzziness of their categorial space. Upon the whole, ‘tipo’ exhibits the basic properties of grammaticalization and is coherent with the cline of the development of discourse particles (Traugott 1995, 2008).
‘Grammaticalization in the present -- The changes of modern Swedish ‘typ’’, by Henrik Rosenkvist and Sanna Skärlund, is a corpus-driven study, aiming to account for the reinterpretation of the word ‘typ’ as a preposition and, subsequently, as an adverb and a discourse marker in the 20th century. This taxonomic lexeme appears to have a strong tendency towards grammaticalization, confirmed by analogous patterns in such languages as Italian, Russian, English and French. The authors conclude that the development of non-nominal uses of ‘typ’ started about 1930 in technology-related discourse, with a high degree of probability, specifically connected with Swedish aircraft.
The papers in the third part, “Gradualness in language change”, explore to what extent diachronic change is gradual, providing case studies regarding particular situations of gradual change, such as language contact.
In “Gradualness in change in English (augmented) absolutes”, by Nikki van de Pol and Hubert Cuyckens, the history of absolute constructions (ACs) is traced from Old English to the present time and analyzed in a constructional perspective. Two major subtypes of ACs are singled out: augmented absolutes, i.e. those introduced by a preposition, and unaugmented absolutes, which do not exhibit any overt marker of syntactic linkage with the matrix clause. In Old English ACs, both the subject and the participle were marked for dative; however, due to the consequent loss of case morphology the constructions became, on the one hand, less recognizable, favoring explicit marking of the beginning of an AC, and, on the other hand, their structural possibilities broadened, such that more and more predicate types entered ACs. Augmented absolutes, introduced by the preposition ‘with’, initially indicated manner or a sort of an accompanying circumstance. Their frequency gradually increased over time, ousting other augmentors (e.g., ‘after’, ‘at’, ‘upon’, ‘by reason of’, common in Middle and Modern English), their meaning evolved towards vagueness and generality, and ‘with’ underwent semantic bleaching, becoming a lexically empty marker of the beginning of an AC.
In “Grammatical encoding of referentiality in the history of Hungarian”, by Barbara Egedi, the development of the definite article from a distal demonstrative modifier is explained within a Minimalist framework. The paper presents a piece of research from a huge project on the generative diachronic syntax of Hungarian. The author assumes that as early as in Old Hungarian this element was already a fully-fledged article, with a structural position of its own; hence, it encoded definiteness at the syntactic level, notwithstanding the coexistence with the homophonous demonstrative marker and, initially, a more limited distribution with respect to the contemporary language. A parallel is drawn with the rise of the definite article out of the Latin demonstrative ‘ille’, originally located in the specifier of the Determiner Phrase and, together with the loss of the first syllable, was reinterpreted as an element in D (Giusti 2001).
“Gradualness in contact-induced constructional replication: The Abstract Possession construction in the Circum-Mediterranean area”, by Chiara Fedriani, Gianguido Manzelli and Paolo Ramat, surveys the diachronic spread and synchronic areal distribution of the Abstract Possession ‘habēre’-construction, which appears to originate in Classical and Late Latin and to be grammaticalized and diffused to different extents across a wide range of Circum-Mediterranean languages. Generally speaking, there is a strong semantic bond between Possessors and Experiencers in the languages of the world, and constructions conveying concrete possession are frequently recruited to express feelings and states, for instance, being right/wrong, or even the age of a person. The present study takes into account 16 concepts pertaining to several semantic domains (physical feelings and types of pain, emotions, moral states, etc.) in a sample of languages which comprises Albanian, Bulgarian, Egyptian Arabic, French, Italian, Macedonian, Maltese, Modern Greek, Moroccan Arabic, Palestinian Arabic, Serbo-Croat, Slovenian, Spanish, and Turkish. The authors come to the conclusion that physical feelings are most compatible with the AP construction and establish a semantic hierarchy (‘physical feelings > mental feelings’), which, actually, implies that a language which uses the abovementioned construction to convey a mental feeling is prone to use the same strategy for physical feelings, whereas the opposite implication does not hold.
In “‘Binding Hierarchy’ and peculiarities of the verb ‘potere’ in some Southern Calabrian varieties”, by Alessandro De Angelis, a range of constructions with the modal verb ‘potere’ (‘can, be able to’) is accounted for in terms of syntax, semantics and contact-induced grammaticalization in several Calabrian varieties which have been in contact with some Greek dialects spoken in Southern Italy. In particular, the author demonstrates how the infinitive in dependent clauses was gradually replaced by finite verbs introduced by a complementizer deriving from Latin ‘mŏdo’ or ‘quod’. De Angelis argues that the process started in purposive and completive sentences with desiderative predicates, extending later on to all the dependent clauses with irrealis semantics. Since a very similar path of change is attested in Late Greek and in the Greek varieties of Italy, it can be considered a case of replica grammaticalization. The final step in the process is constituted by the modal verb ‘potere’, which begins to take finite forms of dependent verbs, although it is not possible in the Greek dialects of Italy, where the analogous verb is the only exception to the construction.
The volume under review is certainly an important contribution to the study of the interaction between synchrony and diachrony. It will be of interest to a wide range of linguists working on grammaticalization, language change, typology, areal linguistics and dialectology.
The very term ‘interface’, in the title, is used in a non-trivial way. In the literature, it is conventionally used to refer to the interaction between different linguistic levels, conceived of as modules. As a matter of fact, synchrony and diachrony have nothing to do with linguistic levels, or modules; they are two domains of linguistic research.
Although the editors state that the focus of the volume is on theoretical and methodological issues (p. 1), the volume is more empirically rather than theoretically driven, which, I suppose, is determined by the subject itself. Even the fact that every article ends with a short theoretical and methodological appendix, titled “Focus on the dynamic interface between synchrony and diachrony”, does not counterbalance the empirical tendency. It looks more like a collection of case studies illustrating some theoretical claims concerning diachrony and its relationship with synchrony, usually points well-established in literature. The exceptions are the editors’ introductory chapter and the paper by van der Auwera, where the methodological/theoretical focus is really central. The papers by Trousdale and Currie, who provide a thorough analysis of gradualness in language change as a general problem, as well as Fedriani, Manzelli & Ramat with their interesting reflections on contact-induced grammaticalization are also notable in this respect.
Rosenkvist and Skärlund strongly criticize proponents of Grammaticalization Theory for the extensive use of data from languages without written records from earlier stages, in particular Bybee et al. (1994) and Heine & Kuteva (2002). The reconstruction of morphosyntactic change in such cases presents well-known methodological problems, but to assert that, in principle, there is no such a thing as a reliable reconstructed path of grammaticalization and that “reconstruction is no more than guessing” (p. 334) is too drastic. Reconstructions rely on the Comparative Method, not mere guessing. Reconstructions are needed in diachronic typology to provide balanced language samples, typically including linguistic families lacking written records from earlier stages. As for Heine and Kuteva, they are always explicit about reconstructed vs. attested instances of language change in their lexicon.
The paper by Fedriani, Manzelli and Ramat offers numerous insights into contact-induced grammaticalization phenomena and lexical typology, but some of their data seems inaccurate, as it is not clear how the authors deal with synonymy and synchronic constructional variation. Considering the focus of the present volume, it is somewhat surprising that the corresponding constructions from the languages under comparison are reported as if they were the only strategies available. For instance, the authors claim that “the two intersubjective states of “being right/wrong” constitute a lexical island … as they are always coded by means of prototypical possessive constructions in our data, the only exception for “being right” being Macedonian …” (p. 411). However, in both Bulgarian and Macedonian, constructions with the verb ‘to be’ are used to convey the concepts ‘to be right’ and ‘to be wrong’ along with the possessive constructions. In Bulgarian the meaning ‘to be right’ is conveyed by the verb ‘to be’ combined with a predicative adjective, ‘prav săm’ (lit. ‘right be.1SG’), as well as by the possessive construction ‘imam pravo’ (lit. ‘have.1SG right’), meaning also ‘to have the right (e.g., to do sth)’. The same holds for Macedonian, where the “existential” construction ‘vo pravo sum’ (in right be.1SG) coexists with the possessive ‘imam pravo’ (‘have.1SG right’). In Bulgarian, the concept ‘to be afraid’ can be expressed by the stative verb ‘straxuvam se’, the stative periphrasis ‘strax me e’ (fear 1SG.ACC be.3SG) and, finally, by the ‘habēre’-construction ‘imam straxa’ (have.1SG fear.DEF); ‘to be worried’ corresponds not only to ‘raztrevožen săm’ (‘worried be.1SG’), but also to ‘imam griži’ (‘have.1sg worry.PL’) and ‘griža me e (za teb)’ (‘worry 1SG.ACC be.3SG (for you.ACC)’). Structural variation can also be observed in the semantic field of pain: En. ‘to have a stomach ache’ corresponds not only to Blg. ‘imam stomašni bolki’ (have.1SG stomach.ADJ.PL aches.PL), but also to ‘boli me stomaxăt’ (ache.3SG 1SG.ACC stomach.DEF); ‘to have a headache’ corresponds both to ‘boli me glavata’ (ache.3SG 1SG.ACC head.DEF) and ‘imam glavobolie’ (have.1SG headache). If this kind of systematic variation is not taken into account, the quantitative analysis will be skewed.
On another minor point, because topics across some papers intersect (e.g., grammaticalization of the noun ‘type’ in different languages addressed by Voghera and Rosenkvist & Skärlund, or accounts of the category of gender in Dutch by De Vos and Semplicini), the volume would have benefited from cross-references.
Overall, though, the book makes a very good impression and is full of insightful analyses of a wide range of morphosyntactic and lexical phenomena.
Bybee, Joan, Perkins, Revere & Pagliuca, William. 1994. The Evolution of Grammar. Chicago, IL: The University of Chicago Press.
Croft, William 2001. Radical Construction Grammar: Syntactic Theory in Typological Perspective. Oxford: Oxford University Press.
Giusti, Giuliana. 2001. The birth of a functional category: From Latin ILLE to the Romance article and personal pronoun. // Current Studies in Italian Syntax. Essays Offered to Lorenzo Renzi [North Holland New Linguistic Series 59] Guglielmo Cinque & Giampaolo Salvi (eds.), 157-171. Oxford: Elsevier.
Heine, Berndt & Kuteva, Tania. 2002. World Lexicon of Grammaticalization. Cambridge: Cambridge University Press.
Traugott, Elisabeth Closs. 1995. The role of the development of discourse markers in a theory of
grammaticalization, Paper presented at ICHL XII, Manchester.
Traugott, Elizabeth Closs & Trousdale, Graeme. 2008. Gradience, gradualness and grammaticalisation: How do they intersect? // Gradience, Gradualness and Grammaticalisation [Typological Studies in Language 90]. Elizabeth Closs Traugott & Graeme Trousdale (eds.), 19-44. Amsterdam: John Benjamins.
Traugott, Elisabeth Closs. 2010. (Inter)subjectivity and (inter)subjectification: A reassessment. // Subjectification, Intersubjectification and Grammaticalization. Kristin Davidse, Lieven Vandelanotte & Hubert Cuyckens (eds.), 29-71. Berlin: Mouton de Gruyter.
| ABOUT THE REVIEWER:
Anna Alexandrova holds a degree in Russian and English philology. She is now a PhD student in linguistics at the Scuola Normale Superiore di Pisa (Italy). Her research interests include linguistic typology, Aktionsart, aspectual systems and verbal morphology, both in synchrony and diachrony.