Introduction

It is a commonly known fact that language change can be observed at different linguistic levels, which correspond to the traditional disciplines of phonology, morphology, syntax and semantics. In this special issue we concentrate on morphological and syntactic changes in Spanish, although we do not exclude phonological or semantic change, as long as they are linked to or relevant for the discussion of a certain morphological or syntactic evolution.

Whereas grammatical change for many decades has been an important issue in the work of a number of Spanish historical linguists, the interest in this kind of change received a strong boost with the development of grammaticalization studies, which undoubtedly have greatly enhanced our knowledge of concrete linguistic evolutionary processes, as well as our understanding of the evolution of grammar in general1.

Work on grammaticalization developed mainly from the late twentieth century onwards. However, it was far from being a novelty then, since various previous studies already touched upon this concept, particularly several nineteenth-century German linguists who dealt with central issues of grammaticalization studies. The term itself was coined by Meillet in 1912 (Meillet 1912), and Kuryłowicz returned to the concept and the process of grammaticalization in 1965 (Kuryłowicz 1965). A few years later, Givón (1971) elaborated further on the topic, but it was not until 1982 that the first monograph on grammaticalization was conceived by Lehmann (only published in (Lehmann 1995); for more information see Narrog and Heine 2011). Finally, the decade of the nineties meant the consolidation of grammaticalization as an approach for studying grammatical change. It provided a framework that made it possible to systemize the study of grammatical change, understood as the evolution from a lexical item or construction to a grammatical word or as the transition from a grammatical item to a more grammatical item.

Unquestionably, numerous processes of grammatical change fit in this definition. However, others, such as the development of the preterite subjunctive in Spanish or the emergence of prepositions such as Old Spanish no obstante (‘in spite of’) or no embargante (‘notwithstanding’), fall outside the limits of grammaticalization proper. Therefore, alternative approaches arose in order to explain how these and other changes took place.

Concepts such as exaptation, capitalization, refunctionalization and adfunctionalization emerged as a result of the need to explain certain evolutionary processes that did not match with the evolutionary patterns proposed by grammaticalization theory. For instance, it is not unusual that languages recycle grammatical material for uses different from its original etymological use. That was the case with Latin plural neuter nouns that ended in –a, whose ending in Spanish, instead of being a plural marker, came to be reanalyzed as a singular feminine marker of collective nouns as words like leña prove (from Latin neuter lignum, plural ligna, ‘wood’, hence ‘firewood’ in Spanish; cf. Penny 2004, p. 122). Interestingly, the change that experimented these Latin neuter plural nouns in their transition to Spanish took place at the morphosyntactic as well as the lexical level. However, a grammaticalization path cannot be invoked to explain the reinterpretation of this ending, since it goes beyond the mere reanalysis of endings in –a as feminine markers in Vulgar Latin, as it also entails the retention of the original plural sense in the collective meaning of these words.

In his well-known and much debated article of 1990, Lass launched the notion of exaptation, which refers to “the opportunistic co-optation of a feature whose origin is unrelated or only marginally related to its later use” (Lass 1990, p. 80). Borrowed from evolutionary biology, he used this term to describe a kind of linguistic evolution in which a grammatical element is recycled and redeployed for a new linguistic purpose. In line with Lass’ (1990, 1997) ideas, it may be maintained that in the course of time the plural neuter marker –a had become obsolete, since only the masculine and feminine gender survived as such in Spanish, and plural meaning was expressed by –s. While neuter singular nouns, ending in –um, in most cases became masculine in Spanish (–um > –o), the neuter plural marker in –a had become a sort of morphological junk, a useless piece that could be exploited in order to signal collectivity. Obviously, the fact that feminine nouns in Spanish frequently ended in –a favored the morphological integration of this type of nouns.

Connecting with Lass’ original idea, Pountain (2000, p. 295) proposed the term capitalization, which he defined as “an attempt to label the historical process by which a linguistic feature which already exists in a language comes to be substantially exploited for wider purposes, sometimes simply making overt distinctions which were previously covert, but sometimes apparently creating new expressive possibilities”.

Subsequently, in order to distinguish between instances of exaptation in which the original function had been lost and instances in which a new function was added to the original one, and to clarify the somewhat diffuse concept of exaptation, Smith (2011) introduced two types of change: refunctionalization and adfunctionalization. He categorized the evolution of the Latin plural neuter nouns as a case of adfunctionalization, i.e., “the process in which a form assumes a new function alongside or in addition to its original function” (Smith 2011, p. 305), arguing that the number opposition of singular/plural continued to exist, but was no longer expressed by the original neuter morphology (Smith 2011, p. 296).

More recently, Van de Velde and Norde (2016) have traced the origin and history of the concept of exaptation, showing that ever since its introduction it has been susceptible to divergent interpretations, leading to terminological confusion and controversy. Broadening the scope of the term, the authors conceive exaptation as “the leap-like co-optation of a trait for a new function that is not immediately related to its former function” (Van de Velde and Norde 2016, p. 10). For Van de Velde & Norde the key issue is no longer whether the new function corresponds to an obsolete word or grammatical function, but the fact that the evolution does not follow its predicted or common course. In this way, exaptation has been applied to deal with unexpected changes. In Spanish, examples can be found in the acquiring of a subjunctive value by the pluperfect indicative (cantaveram; Rodríguez Veiga 2006, p. 174 ff.) or in the extension of the definite article at the head of completive that-clauses (Wall and Octavio de Toledo y Huerta 2016) as in El que no sepas caminar y mascar chicle al mismo tiempo no significa que el resto de la raza humana sea como tú. (‘That you do not know how to walk and chew gum at the same time does not mean that the rest of the human race is like you.’).

In spite of the indisputable merits of grammaticalization theory and concepts as exaptation, capitalization, refuncionalization or adfuncionalization, they generally disregard or do not properly allow for other aspects that are highly relevant for certain grammatical changes. As a matter of fact, languages are not only used in specific conversational contexts, but they are also employed in written texts, and as Narrog and Ohori (2011) showed, a grammatical change may also start or develop in this latter modality. In the same vein, even so-called dead languages, as is the case with Latin in the European medieval countries, may generate linguistic innovations, that originated either in written texts or in language employed in formal contexts (cf. Jover 2010). These innovations in Latin were transferred to European vernaculars via written texts and, afterwards, were introduced into the spoken language through formal registers. Since, in our view, these features are an integrated part of certain types of grammatical changes, it is essential to take into account the cultural and social context in which these particular changes occur.

The present special issue examines the usefulness and applicability of a number of the above-mentioned concepts and processes of change to a range of linguistic changes in Spanish and Latin that cannot be easily or can only be partially accounted for within the framework of grammaticalization. Rather than challenging the insights of grammaticalization theory, the different contributions to this special issue demonstrate that exaptation, capitalization, refunctionalization and adfunctionalization, as well as changes motivated by rhetorical guidelines, constitute interesting and valuable notions that allow for a better understanding of specific language changes in Spanish and, by extension, of language change in general.

This special issue consists of three parts that focus on different components of the grammar. Each part comprises a number of case studies on specific grammatical topics in Spanish or Latin. Ideally, the reader should read the papers in the order we establish below, since this order, we believe, renders a thematically coherent overview of the work included in the present issue.

However, every paper may also be read independently and actually, as the reader will find out, the order of the papers on the journal website differs from the one proposed here. Thus, the reader may also play hopscotch, as in Julio Cortázar’s famous novel Rayuela/Hopscotch, and read the papers in his own preferred order, or simply follow the order of the journal website.

Pountain opens the first part of this special issue, which concentrates on changes in the noun phrase. The author argues that the development of the definite and indefinite articles in Spanish may be regarded as an instance of capitalization or adfunctionalization. His meticulous quantitative and qualitative analysis reveals that the definite article is the result of a partial exaptation of the Latin demonstrative ille, whereas the indefinite article still preserves its original value of numeral. However, as Pountain shows, both articles developed additional functions, related to the original ones.

Pato studies the history of the sequence <indefinite article + possessive + noun>, as in un su amigo (literally ‘a his friend’), which existed in Medieval Spanish, fell in disuse in Modern Peninsular Spanish but has survived in some Spanish American dialects, particularly in Central American Spanish, although with a new discursive value. The claim of the author that the construction can be understood as an example of refunctionalization is convincingly grounded in a qualitative analysis of the distinct values of the possessive construction.

The third contribution concerning the noun phrase explores the relationship between refunctionalization and usage frequency. Rosemeyer proposes that folk etymology processes can be characterized as a type of refunctionalization, since in folk etymology obsolescent morphemes are replaced with morphemes that still have a function in language. In order to test this claim, he conducted a questionnaire survey in which participants were asked if they accepted the assumption that a target word derived from a proposed etymon that, however, was false. Rosemeyer’s results indicate that usage frequency plays a role in folk etymology processes and, hence, in refunctionalization.

The second part of the special issue, dedicated to changes in the verb phrase, starts with a contribution by Kailuweit, who studies the use of <to be + past participle> with intransitive verbs in Medieval and Early Modern Spanish. Whereas in Latin <esse + past participle> mainly had an aorist reading, Medieval Spanish <ser + past participle> has to be interpreted as a resultative construction, thus challenging the common assumption of a grammaticalization path from resultative constructions to aorist constructions. In order to describe this process adequately, the author introduces the term decapitalization reversing Pountain’s concept of capitalization.

Hernández Díaz analyzes the diachronic development of the first person plural verb form habemos (literally ‘we have’, now ‘there are’), which originally was a full verb expressing possessive meaning but in the course of time became an auxiliary of perfect tense and in Modern Spanish has an existential value. Hernández Díaz argues that the recycling of habemos can be conceived as a case of refunctionalization, triggered by the fact that it presents communicative advantages that other existential verbs lack.

The topic of part three is the origin and development of a number of discourse markers and grammatical words. In her contribution, Company stresses the necessity to consider context as the unit of language change when grammatical words are involved, being context always the locus of change in this particular cases. Typically, these changes exhibit an active spreading of a form to new contexts, which, however, does not entail a change in category or grammatical status. Evidence for these claims comes from the diachronic expansion of the Spanish preposition a (‘to’).

The next two contributions discuss the evolution of the Spanish preposition mediante (‘by means of, with’) in different historical and diachronic periods. Artigas examines its origins in Late Latin and analyses the different uses of the verb mediare (‘to be in the middle, to mediate’), particularly its present participle in ablative form mediante. This analysis allows her to shed light on the linguistic and extra-linguistic circumstances that accompanied the emergence and behavior of mediante in the first centuries of Spanish. At the same time, her work highlights the more general issue of the importance of discursive traditions and historical-cultural factors in the evolution of linguistic units.

Garachana Camarero studies the origin of the use of the preposition mediante (‘by means of, with’) in Medieval Spanish. She argues that, whereas traditionally its presence has been described as the result of a process of grammaticalization, empirical data extracted from two large digital diachronic corpora support the claim that this preposition is the outcome of a grammatical calque of Latin that entered the Spanish language through Aragonese and Catalan.

Subsequently, Azofra analyzes the role of elision in the evolutionary process of the additive connective aparte (‘besides’) and shows that this particle, that presents serious problems of categorization, as reflected in various lexicological works, introduces an argument that is not indispensable to reach a certain conclusion. In this sense, it is different from all other additive connectors in Spanish. The author uses the case of aparte to draw some more general conclusions on the explanatory force of grammaticalization theory and, at the same time, highlights the importance of context as the locus of change.

In the last contribution to this special issue, Nieuwenhuijsen studies the historical development of the Spanish doublet ante-antes (‘before’) and explores the question whether the adverbial –s, as present in antes, was exploited to mark a semantic difference between the two forms, once the original adverbial value of final –s had become obsolete. Quantitative data are presented to test this claim and to analyze the distribution of the different meanings of the two members of the doublet.

We would like to express our gratitude to the anonymous reviewers of the papers of this issue, whose comments have definitely helped to improve the research that is presented here. Our special thanks go to Christopher J. Pountain for helping with the English editing of several papers.