New Approaches to Spanish Dialectal Grammar: Guest Editor’s Introduction

1. Introduction

If we understand Language as everything that is selected, numbered, merged and sent, and Grammar (both of the language and of the individual) as the union of Lexicon and Syntax, we share the same theoretical vision. Under this framework, we would only have a (generative) Syntax, and that Syntax can be materialized in various ways.

Furthermore, variation is nothing other than the extension of usages to other categories and structures (Bosque 2023, p. 15). Nonetheless, there has not always been agreement over what linguistic variation is and how it is to be accounted for. A rivalry between two somehow complimentary views seems to be at play.

On one side, it has been explained as a permanent syntactic diglossia (Kroch 2001), i.e., as an effect of the coexistence of grammars. Today it is believed that speakers do not have different grammars, but an open range of lexical items, with some of them triggering syntactic effects. In formal frameworks, variation has been held to boil down to differences in the array of features of specific lexical items (Chomsky 1981, 1995; Borer 1984; Baker 2008; Adger and Smith 2010; Eguren 2014; Mare 2023). Grammatical variation, thus, stems from the feature composition of items stored in the Lexicon, which are then assembled in Syntax. This thesis has come to be known as the Borer–Chomsky Conjecture. Following this conjecture, all parameters of variation are attributable to differences in the traits of particular items in the Lexicon. In addition, the dialectal system is not static but rather participates in the same process of change as the general/standard language (Szmrecsanyi and Kortmann 2009; Szmrecsanyi and Röthlisberger 2019). For all these reasons, variation regards how the information is materialized and is explained by a grammatical principle.

On the other hand, language is composed of its varieties. Based on when and how Spanish is acquired (first-language acquisition or second-language acquisition) and contact language, Spanish speakers can be divided into (i) native speakers and (ii) non-native speakers, with a distinction between (ii.a) second-language speakers and (ii.b) foreign-language speakers. A basic typology of different types of varieties (see Crystal 2004) of Spanish language may be as follows:

• Native L1 varieties. Speakers have an intranational speech community. There is one international Spanish and 20 “colonial” varieties.
• Native L1 varieties in contact with other varieties of different (Castilian) Spanish dialects.
• Native L1 varieties in contact with other European languages (e.g., English, French and Portuguese).
• Bilingual varieties. Bilingual speakers learn Spanish as an additional language on top of their mother tongue or symmetrically (e.g., Catalan–Valencian, Galician and Basque).
• Indigenized L2 varieties. These varieties are in contact with indigenous languages in America (Nahuatl, Maya, Quechua, Aimara, Guarani, Mapudungun, etc.).
• Pidgin and creole languages based on a particular lexifier (e.g., Chabacano, Palenquero, Jopará, Bozal and Papiamento).

Nowadays, all speakers develop their own Spanish variety and are creative users of Spanish with their possible “deviations” from the standard norm. For their part, non-native speakers are “norm-dependent” on native speakers and prefer the “school-variety” of Spanish. As Ranta (2022) shows, the difference is the “exposure”, “model” and “production”. Exposure refers to all the (standard or non-standard) Spanish input native speakers receive in real-world settings and non-native speakers read and listen to in classroom settings. Model refers to the kind of (endonormative) Spanish the speakers are supposed to imitate for a usefully speech community integration in each Spanish speaking country. In addition, production is the outcome that speakers produce themselves. In practical terms, if one speaker is exposed to many different kinds of Spanish varieties, they will better perform.

The problem is still that the same non-standard linguistic constructions, especially but not only in grammar, in both written and spoken language, are known and treated as “innovations” in native speakers but “errors/mistakes” in non-native speakers. Learners are very often (auto) expected to be much more “standard-like” than native speakers are. However, it seems that cognitive processes and linguistic strategies are bound to be similar and shared for all speaker groups (Paradis 2004; Ranta 2022). In this Special Issue, speakers are seen as users (not learners) of Spanish language, and grammatical “deviations” will be considered dialectal features, that is, features without a norm-dependency.

Finally, following Trudgill (2010), in dialect contact scenarios, the majority of language users are speakers who select various features from the heterogeneous input of the feature pool. Plausibly then, childhood language acquisition results in an increase in linguistic variants and complexification. Language contact scenarios entail adult users acquiring a second language, which inevitably leads to simplification because of the limited language acquisition abilities of adult speakers. Lastly, we can predict more simplification in indigenized L2 varieties compared to colonial L1 varieties.

2. Vernacular Universals and More

Following Chambers (2004, p. 128), we understand vernacular universals as “a small number of phonological and grammatical processes [that] recur in vernaculars wherever they are spoken” in the working class, rural vernaculars, pidgins, creoles and interlanguage varieties, in other words, linguistic varieties that lie far beyond normative- or prestige-related pressure. This conclusion follows from the observation that no matter where in the world the vernaculars are spoken, these features inevitably occur.

Vernacular universals are also primitive features, not learned. As such, they belong to the language faculty. They cannot be merely English (or Spanish). They must have counterparts in the other languages of the world that are demonstrably the outgrowths of the same rules and representations in the bioprogram.

In addition, vernacular universals arise in the context of sociolinguistic dialectology as generalizations about intralinguistic variation (so far, mainly from English dialects), but their universal status is emerging from analyses of putative crosslinguistic counterparts.

This notion of vernacular universals is interesting from a typological, functionalist-cognitive and formalist point of view. Conceptually, the notion gives rise to a number of critical questions. However, vernacular universals can be found in non-standard varieties of English, for example, Angloversals proper. As Kortmann and Szmrecsanyi (2007) pointed out, vernacular universals add a crucial social dimension to research in crosslinguistic variation and language universals.

The criteria for universalhood are as follows:

• The candidate feature should be attested in a vast majority (>80–90%) of a given language’s vernacular varieties.
• The candidate feature should not be patterned geographically or according to variety type (L1, L2, or pidgins/creoles).
• For the sake of crosslinguistic validity, the candidate feature should not be tied to a given language’s typological make-up: inflectional (e.g., Hungarian), isolating (Vietnamese), etc. If it is, vernacular typoversal is the more appropriate term.
• The candidate feature should be crosslinguistically attested in a significant number of the world’s languages (especially in those without a literary/philosophical tradition).

As far as the Spanish language is concerned, varieties of this language share common (non-standard) features, both in space and in time (Pato 2023):

• They can be registered in all kinds of varieties: rural dialects, children’s speech, creole languages and the interlanguage of non-native speakers (hence their “universality”).
• They are documented in most vernacular varieties (although with different percentages).
• They are not modeled geographically (diatopically) or according to the type of variety.
• They have interlinguistic validity, and they are not linked to the typology of Spanish (inflection not exclusive to this language).
• They appear in other languages at the interlingual level.
• They arise naturally, as a result of basic principles of language processing, and are not learned (they depend on the input), compared to many standard traits that are usually the result of standardization or selection and are formally learned.
• They can serve as indicators of linguistic diffusion and (social) extension.
• In some cases, they can overcome the “transfer/retention” dichotomy, as well as the “conservation/innovation” dichotomy, as they are old and modern forms at the same time.
• Between vernacular universals and contact-induced change, there would be a continuum; not everything has to be a “vernacular universal” or “contact-induced change”.

Vernacular universals have to be spread in different varieties and used in the same way, but the quantitative aspect does not play a prominent role. A small number of speakers only use some dialectal features. Spanish grammatical constructions that deviate from the standard but occur across native and non-native Spanish speakers’ production (Kortmann 2004) are very frequent.

Regarding the typology of -versals, Szmrecsanyi and Kortmann (2009, p. 33) propose an analytical distinction between them:

• Vernacular universals: features that are common to spoken vernaculars.
• Areoversals: features common in languages or language varieties which are in geographical proximity.
• L(anguage)-versals: features that tend to recur in vernacular varieties of a specific language, such as (vernacular) Angloversals, Francoversals, etc.
• Varioversals: features recurrent in language varieties with a similar sociohistory, historical depth and mode of acquisition.
• Typoversals: features that are common to languages of a specific typological type.
• Phyloversals: features that are shared by a family of genetically related languages.

The term Angloversals cover non-standard features that occur in both native and non-native production. I propose to use the term Hispanoversals for the same situation in the Spanish language. We define Hispanoversals as being recurrent features in all varieties and types of Spanish.

In addition, large-scale quantitative studies on morphosyntactic variation show that variety type (L1, L2 and pidgins/creoles) is more important than world region (areoversals). Moreover, lexical and morphological transparency entails analytic structures where “the relation between form and meaning is as transparent as possible” and “every single meaning is expressed in a separate form” (Kusters 2003, p. 21). Simplification leads to increased analyticity, and complexification leads to increased syntheticity.

3. Dialectal Variation in Spanish: Some Features

As we saw earlier, if we take the Borer–Chomsky Conjecture to be accurate, dialectal variation is attributable to differences in the features of specific items in the Lexicon. In view of the above, if speakers of the vernacular varieties have the same cognitive background as speakers of the standard variety, it is important to extend this hypothesis to the dialect data and give those data the treatment they deserve (Di Tullio and Pato 2022). Furthermore, the limits between the vernacular and the colloquial features are not precise. Some phenomena are vernacular in all dialects, and other phenomena are vernacular only in some areas.

Here, we have some cross-varietal features in Spanish:

• Conjugation regularization, or the levelling of irregular verb forms;
• Default singulars, or the subject–verb non-concord;
• Multiple negation, or negative concord;
• Copula absence, or copula deletion;
• Lack of inversion in main clause sí/no questions;
• Yo/tú instead of mí/ti in coordinate subjects;
• Adverbs taking the same form as adjectives;
• Absence of plural marking after measure nouns;
• Lack of inversion and lack of auxiliaries in qu- questions;
• Degree modifier adverbs lack -mente;
• Special forms or phrases for the second-person plural pronoun;
• Levelling of difference between tenses;
• Double comparatives and superlatives;
• Irregular use of articles;
• Levelling of preterit/past participle verb forms: regularization of irregular verb paradigm;
• Lack of number distinction in reflexives.

As we can see, some linguistic features are particularly diagnostic of specific types. In addition, the Fundamental Dialectology Principle (Nerbonne and Kleiweg 2007, p. 154) establishes that geographic proximity between dialects or varieties should predict the linguistic similarity between these dialects and varieties. It is understood that geographic proximity does not always mean linguistic similarity, as shown by not-so-similar linguistic varieties that share a geographic space. The reverse situation also holds: geographic proximity is by no means a sufficient condition for linguistic resemblance. We found an under-representation of “rare” morphosyntactic features in previous works. First, not everything widespread in one particular country, region or variety type is necessarily widespread in other varieties of Spanish. Second, not everything recurrent in varieties of Spanish is by definition universal crosslinguistically. The following are some features that are catalogued according to feature groups:

• Pronouns, pronoun exchange and pronominal gender;
• NP;
• Verb phrase: tense and aspect;
• Modals;
• Verb morphology;
• Adverbs;
• Negation;
• Agreement;
• Relativization;
• Complementation: “para + infinitive”; comparative clauses “contra más”;
• Discourse organization and word order.

The type of linguistic features, the kind of data sampled and the dialect area examined play a crucial role in determining the importance of geography for the classification of dialects.

4. On Late Insertion

According to the Strong Modularity Thesis, Syntax does not contain all the features of forms, but only the elements relevant to the application of syntactic operations. Idiosyncratic properties (e.g., that querer is a verb ending in -er, its meaning is close to “amar” and it begins with /k-/) are listed in the (strict) Lexicon, which is accessed post-syntactically in late insertion models. The theoretical burden of these minimalist hypotheses lies, thus, not in the over-specification of lexical items or in ad hoc explanations of particular constructions but in isolating more abstract, minimal principles of syntax that account for all possible human grammars. In late minimalist models precisely inquiry into what is the nature of these syntactic principles, and which are they (i.e., Merge, Move, Copy, Agree, Concord, etc.) has figured at the core of linguistic theory.

In the Distributed Morphology model (Halle and Marantz 1993; Mare 2023), Syntax builds structures by joining (assembling) abstract features and sends them to be interpreted (encyclopedic knowledge) and pronounced (phonological–phonetic information).

In phonology, the syntactic–semantic features are interpreted by searching in a second lexical list (the Vocabulary) which phonological exponent the language has to materialize those features. Moreover, since these phonological exponents are not in the syntactic derivation, it is said that there is a late insertion.

Morphological segmentation only takes place in the phonological component branch. These operations affect morphophonology and not syntax because when they are applied, the form has already been completed (they are not in Syntax). They do not affect semantics either because semantics interprets the syntactic features in the semantic interpretative component (in another branch, see Figure 1 below).

The variation does not imply any change in the Syntax; the features are universal, and the combination mechanisms are the same. What changes are the options for materializing the information. A rationale for the explanation of linguistic variation should not lead us to assume that the loci of variation are the abstract syntactic principles that merge lexical items into subsequently larger phrases. It seems plausible (and even promising) to assume that linguistic variation does not alter the abstract principles of grammar but rather more language-dependent facts, such as the feature specification of items in the Lexicon, the level at which some abstract operation applies, and so on. The Syntax generates a structure which is categorized and transferred. In the Logical Form (LF0), the interpretation is checked (encyclopedia), and in the Phonological Form (PF), it is externalized (materialized). The differences would be encoded in different structures within the lexical entries of the words. However, phonological exponents can be added later in the derivation. Therefore, the structure created by Syntax may be modified before the insertion of phonological exponents. Moreover, this type of change takes place in the component called Morphology and depends both on the specific properties of the language and on the inventory of vocabulary items.

We think that this model can explain all the variations, including the differences between child and adult grammars, between grammars in contact situations, between diachronic and synchronic grammars, and between varieties/dialects and the standard language.

Traditionally, word formation and sentence formation take place in the following distinct components: the combination of morphemes to form words, in the Lexicon; and the combination of words to form sentences, in the Syntax. In the Distributed Morphology model (Mare 2015, 2023), everything is Syntax, including the following: a single component (the Syntax) that can generate words and sentences, and post-syntactic operations (which do not affect interpretation). Moreover, the Lexicon is distributed, which is to say the information is listed and located in different places in the structure of the grammar:

• List 1: Abstract features, roots and categorizers (nominal, verbal and adjectival), without phonological content and proper to a Language or Variety ([Plural], [Human], [Future]). Syntax does not manipulate phonological material. Located in Syntax.
• List 2: Vocabulary items. Allows abstract information [features] to be materialized/pronounced (with phonological content) in the nuclei of syntactic projections (or nodes, via the Subset Principle). Located in the Phonological Form.
• List 3: Encyclopedia. Information related to meaning and interpretation (in a given context). Located in the Logical Form.

As we have said, between the structure, Syntax, and its materialization, there is a component called Morphology. Here, the post-syntactic operations take place that only affect the externalization/pronunciation (but not the interpretation) and can alter the conditions of the insertion of vocabulary items. Of course, this is only one of the models of the formal framework.

5. Individual Contributions to This Special Issue

This Special Issue aims to present and outline—from different theoretical frameworks—an array of dialectal grammatical phenomena described and analyzed in articles that study the way in which words are combined and the meanings to which these combinations give rise (Bosque and Gutiérrez-Rexach 2009). This Special Issue will constitute a useful supplement to the existing literature on the topic. The contributions concentrate on non-standard grammatical features, oral data, rural speakers, dialect contact, and vernacular universals in Spanish. They are listed in order of publication.

1

María Mare. The author examines the distribution of the “genitive” pronoun in non-nominal domains (i.e., adverbial, verbal and adjectival) from a neo-constructionist approach (the late insertion of phonological exponents) and explains the complementary distribution between agreement/nominal morphology. Both the synthetic (nuestro “our”) ant the analytic (de nosotros “of us”) options may have the same syntactic structure; the difference is the nature of the nominalizer’s phi features. When the nominalizer cannot value its features, a “non-genitive” pronoun (nosotros) lexicalizes the pronominal structure, and the head p/Place is lexicalized by de.

2

Bruno Camus Bergareche. The author presents a detailed description of some morphological and syntactical phenomena distributed in Castile–La Mancha that are actually documented in a general way (leísmo and laísmo; deísmo; non-standard clitic sequences me se; transitive use of entrar, caer and quedar; reduced desinence for second person plural verb forms like cogís instead of cogéis; etc.) and that show its internal boundaries. This dialectal description helps to know this variety better and not to classify it simply as “transitional” between northern and southern Spanish.

3

Luis Eguren and Cristina Sánchez López. The “Expletive” mismo (“same”) is a non-comparative emphatic use of this prenominal adjective in many American Spanish varieties (Finalmente, Laura se sentó y aceptó el cigarrillo, mismo que nunca encendió “Laura finally sat down and accepted the cigarette, same that she never lighted up”). The authors provide its geographical distribution and a novel analysis of its properties. They conclude that mismo functions as an anaphoric “reinforcer” preceded by a null definite determiner and combines with an empty noun that takes a restrictive relative clause (que).

4

Teresa María Rodríguez Ramalle. This paper studies the variation found in some discursive structures and compares the use of que with evidential adverbs (naturalmente “naturally”, ciertamente “certainly”) and affirmative adverb sí (que) in Latin American Spanish. The proposal is that evidential adverbs are examples of recomplementation.

5

Elena Felíu Arquiola. This paper studies a syntactic construction with a verb in a non-pronominal intransitive variant (congela “freeze”) with property interpretation (La masa de pizza congela perfectamente “Pizza dough freezes perfectly”, instead of se congela). The author pays special attention to the type of verb and establishes a comparison with mediopassive and anticausative constructions. Linguistic factors (the presence of a manner adverb and negation) as well as extralinguistic factors (the type of text and geographical distribution) are also reviewed.

6

José Silva Garcés and Gonzalo Espinosa. They describe the prosodic, semantic and morphosyntactic behaviour of duplicated verbs surrounding an XP that bears the nuclear accent of the phrase (Se fueron por Bariloche, se fueron “They went by Bariloche, they went”) in Patagonian Spanish and propose a biclausal analysis in which each duplicate originates in a different clause (CP1 and CP2). In this case, the structure moves to the left periphery of CP2.

7

Florencio del Barrio de la Rosa. Spanish nouns vary depending on their denotation (niño “male child”/niña “female child”), and when lacking an inflectional cue, they could assume both genders (azúcar moreno “brown sugar”/azúcar blanquilla “white sugar”). This study addresses the socio-geographical factors conditioning gender assignment in European Spanish and shows how gender values are determined and diffused across rural and urban varieties.

8

Sara Gómez Seibane. Clitic doubling (i.e., the co-appearance of a clitic and a correlative syntagma) is an internal/external language interface phenomenon. This study determines the frequency of this use and analyses the following features: definiteness, specificity, the cliticization of direct object, and the accessibility of IO referents in the minds of the speakers.

9

Juan José Arias. This study, within the framework of Distributed Morphology, explores the exclamative use of the feminine definite article la (¡La de chicos que besé en la fiesta! “How many guys I kissed at the party!”). According to the author, this kind of example is not CP but indefinite DP with an exclamative behaviour, which contain a pseudo-relative clause with que. The DP projections (FocP and FinP) allow la to move to Spec-FocP and be interpreted as an exclamative operator.

10

Edita Gutiérrez-Rodríguez and Pilar Pérez Ocón. This paper offers a formal analysis for deísmo (i.e., the insertion of preposition de (“of”) before an infinitive clause: Me apetece de salir “I want to go out”). The authors propose that this preposition (de) is located in a projection below that of the de in dequeísmo constructions (Creo de que hay tiempo suficiente “I believe there is enough time”). Moreover, deísmo is lexically associated with certain verbs but not with a semantic class. Finally, they propose that there is not an evidential meaning associated with deísmo.

11

Jorge Agulló. Unlike Catalan, existential constructions in Spanish are sensitive to definiteness or quantification restriction. This fact prevents personal pronouns (*Había él en la habitación “There was him in the room”), proper nouns (*Hay Inés en la habitación “There is Inés in the room”) and definite constituents (*Hay tu libro en la habitación “There is your book in the room”) from occupying the pivot position. In contrast, contact varieties between Spanish and Catalan show a range of examples not yet analyzed. The author quantitatively studies the variation between definite and indefinite pivots in both languages, and offers a new explanation based on the person, number and gender (phi features) of the pivot (haber).

12

Avel·lina Suñer Gratacós. This paper, using a compositional approach, analyzes immediate succession subordinators in temporal subordinates, its different materialized forms and its patterns. She concludes that new subordinators can be created by adding a component [immediacy] to a previously temporal subordinator.

13

Aldo Olate and Ricardo Pineda. The authors analyze and compare the nominal possessive constructions (Su casa de Pedro “His house of Pedro”) in bilingual Mapudungun/rural Spanish and monolingual rural and urban Spanish. They show this use of copying as a strategy of the speakers and its communicative needs.

14

Silvia Gumiel-Molina, Norberto Moreno-Quibén and Isabel Pérez-Jiménez. In innovative uses of estar (Voy a visitar usuarios que están muy morosos “I am going to visit users that are defaulting debtors”), there is no comparison between stages or counterparts of the subject (usuarios) and the property expressed by the adjective (morosos). This study provides a complete characterization of this use, shows its geographical distribution and reviews the lexical–syntactic classes of adjectives that appear in it.

15

Grant Armstrong. Yucatec Spanish exhibits a syntactic construction between an impersonal and passive (Te castigaron por mi tío “You were punished by my uncle”), where the preposition por introduces an agent, the verb is third person plural (castigaron) and an accusative clitic (te) is possible. Labeled as Active–Passive hybrid, the author formally analyzes these hybrid constructions and arges that they are instances of grammatical object passives. The emergence of a null pronoun in the specifier of Voice restricts an agent argument and allows for this type of sentence. In addition, language contact may have played a role in this case.

16

Ignacio Bosque. This author analyzes four vernacular interpretations of the adverb siempre (“always”): (i) continuative, (ii) progressive–comparative (in Rioplatense Spanish), (iii) concessive–adversative (in Mexico and Central America) and (iv) attenuated (in Andean Spanish). He also offers an interpretation of this adverb as a universal quantifier with different semantic nature.

6. Final Remarks

As is known, dialectal grammar (i.e., variation) is fundamental for the description of linguistic areas, but also for the theoretical and typological study of languages. There is variation in all grammatical categories and in many syntactic constructions. Under this precept, we believe that the “problem” of variation depends on how widespread the feature is in the language, and its impact on the standard language.

In addition, from a theoretical point of view, Grammar (of the language/individual) is the sum of the Lexicon and Syntax. We only have one Syntax, and that Syntax can be materialized in different ways or options. In this regard, the syntactic structures (standard and vernacular) would be the same, and what changes are the syntactic–semantic features that cause a superficial distinction between them. If a speaker uses a feature, it is because the linguistic system allows it. As linguists and grammarians, we must check how these usages fit within the language system. If they have not been described previously, we must now think about how Grammar should be extended to include them (Di Tullio and Pato 2022; Bosque 2023; Pato 2023).

Finally, where appropriate, a distinction can be made between vernacular usages (dialectal, “horizontal” features) and vernacular usages (vulgar, “vertical” features). However, if they are always seen and understood in a dichotomous way, neither of them ever enter into Grammars nor are they studied (some of them because they are dialectal, and others because they are vulgar and are thus always left out). In this Special Issue, the authors show us some of them.