Full Title: Panel at LIV Simposio de la Sociedad Española de Lingüística: "Ellipsis and the Architecture of Language: When and Why Can an Element Be Elided (2nd ed.)"
Short Title: LIV Simposio SEL

Date: 26-Jan-2026 - 29-Jan-2026
Location: Madrid, Spain
Contact Person: Carlos Martínez-García
Web Site: https://www.sel.edu.es/liv-simposio-madrid-2026/

Linguistic Field(s): Linguistic Theories; Pragmatics; Semantics; Syntax
Subject Language(s): English (eng)
Spanish (spa)
Language Family(ies): Romance

Call Deadline: 30-Sep-2025

2nd Call for Papers:

Description:

The goal of this panel is to maintain the meeting space for linguists working on ellipsis that was initiated at the LII Simposio de la SEL, held in Madrid in 2024. Given that the panel was successful in 2024 (researchers from different universities and countries, such as Spain, Germany, the United States and Poland attended), we believe that the session can provide a fruitful meeting space at the LIV Simposio de la SEL.

Ellipsis is the anaphoric phenomenon that has aroused most interest in formal linguistics, within generative grammar and outside it (Pollard & Sag 1994; Dalrymple 1999; Merchant 2001). By the term ellipsis, we refer to those cases in which a part of a sentence can be interpreted semantically by virtue of a discourse antecedent, even though it lacks phonetic content. Since the syntactic component systematically associates sound and meaning, the challenge posed by ellipsis lies in explaining why the absence of phonetic content does not entail a loss of meaning. In other words, how can meaning be generated without sound? These questions were raised in the first edition of this panel, and continue to be of scientific interest today.

Although some authors have argued that the semantic content of ellipsis can be recovered without the aid of syntax (Culicover & Jackendoff 2005), empirical evidence points to the contrary. The fact that unpronounced material is accessible to syntactic operations, such as extraction, agreement, or quantificational scope, provides a strong argument in favor of the hypothesis that, in cases of ellipsis, what is unpronounced has an abstract syntactic representation (Johnson 2001; Merchant 2001; Saab 2021). However, this type of argument does not yield the same results for all types of ellipsis. This has led to an intense debate about which ellipsis phenomena have syntactic structure (Hankamer & Sag 1976).

Ellipsis phenomena have been classified into two types. We refer to argument ellipsis and non-argumental ellipsis. In argument ellipsis, nominal arguments of the verb are omitted, as in (1B). In non-argumental ellipsis, the elided material does not represent a syntactically homogeneous class: the elision may scope a finite clause, as in (2a), or be circumscribed to a smaller sentence. For example, in (2b), the elided part is a constituent of the SD, while in (2c) the head of the SV is omitted (the parentheses indicate the elided part of the speech, and the hooks indicate what serves as its antecedent).

(1) Argument ellipsis
A: ¿Ha comprado [Juan] [libros]?
‘Did John buy books?’
B: Sí, (Juan) ha comprado (libros).
‘Yes, (John) has brought (books).’

(2) Non-argument ellipsis
a. Juan [quiere trabajar], pero no sabe de qué (quiere trabajar). [sluicing]
‘John wants to work, but (he) does not know on what (he wants to work).’
b. El [libro] de Cervantes y el (libro) de Quevedo están ahí. [elipsis nominal]
‘The book by Cervantes and the (book) by Quevedo are here.’
c. Juan [comprará] el queso y Pedro (comprará) el vino. [gapping]
‘John will buy the cheese and Peter (will buy) the wine.’

The study of these two types of ellipses offers a unique opportunity to shed light on key aspects of human language architecture. How does the computational component interact with each interface of external systems? What constraints does the computational component impose on syntactic representations? What constraints affecting syntactic representations are instead imposed by the interfaces? What structure do the interfaces with which the computational component interacts possess?

Goals:
This panel is open to papers that address any of the basic questions raised by the study of ellipsis, such as the following, among others.

I. What can be elided?
According to its semantic interpretation, Standard Spanish distinguishes two types of null arguments: definite and indefinite. Definite ones are mostly licensed in subject position of finite sentences, and receive a definite interpretation (Chomsky 1981; Brucart 1987; Fernández Soriano 1989; Holmberg 2005; Camacho 2013). Indefinite ones are mostly licensed in object position, and inherit their indefinite interpretation from a non-specific, indefinite antecedent, usually a bare noun (Campos 1986; Brucart 1999; Laca 2013; Verdecchia 2022). Do the syntactic and semantic differences exhibited by these two kinds of null arguments justify a separate analysis for both phenomena, or can they be explained under a unified analysis (Duguine 2014)? What other types of argument ellipsis exist in Spanish whose syntactic and semantic properties are incompatible with the pro-drop parameter (Ortega-Santos 2023)?

Contrary to what happens in Standard Peninsular Spanish, certain dialects of Spanish allow definite null arguments in object position. These dialects are in contact with on-Indo-European languages that license definite null objects (Suñer & Yépez 1988 for Quechua-contact Quechua Spanish; Palacios 2000 for Paraguayan Spanish in contact with Guarani; Guijarro-Fuentes et al. 2022 for Basque Spanish in contact with Basque). How can the micro-variation observed in this group of dialects be formalized? What mechanisms explain the transmission of this phenomenon from one language to another? What representation do null arguments receive in the minds of this group of bilingual speakers?

II. When?
Within generative grammar, ellipsis has traditionally been analyzed as the result of a post-syntactic operation, either (i) as an operation in Phonological Form (Tancredi 1977; Chomsky & Lasnik 1993; Merchant 2001), or (ii) as a reconstruction process in Logical Form (Zagona 1988; Lobeck 1995). However, within the Minimalist Program, the opposite idea has been defended: ellipsis as a syntactic operation (Aelbrecht 2010; Baltin 2012). According to this hypothesis, ellipsis renders the elided structure inaccessible for further syntactic operations. Therefore, syntactic movement from the elided material would only be possible if it occurred ‘before’ the ellipsis operation. The main argument in favor of this hypothesis is that, for certain types of ellipsis, movement from the omitted structure is possible in some cases, but not in all. This is the case for some null complement anaphora. While the extraction of a relative pronoun is possible from within the null complement of modal verbs such as poder, deber, or querer (Busquets 2006; Dagnac 2010; Saab 2021), as shown in (3a), argument clitics cannot be extracted from the elided SV (Brucart 1987; Depiante 2000), as seen in (3b).

(3) a. Juan besa a quien puede (besar).
‘John kisses whom (he) can (kiss).’
b. Juan no la puede leer, pero María la puede *(leer).
‘John cannot read it, but Mary can (read) it.’
What other morpho-syntactic processes are blocked by ellipsis? Does the fact that some processes cannot occur in ellipsis contexts help to identify the grammatical level where the ellipsis takes place (Lasnik 1999)?

III. Why?
A syntactic constituent must maintain an identity relation with its antecedent to be elided. Is this relation semantic or syntactic, or both (Fiengo & May 1994; Johnson 2001; Merchant 2001)?
In addition to this identity relationship, what syntactic conditions must be in place to license ellipsis? How do the syntactic conditions that give rise to different types of elliptical constructions vary from language to language? Are there specific parameters for ellipsis or does the observed cross-linguistic variation derive from independent principles?

For full list of references, please see the event website.

Guidelines:
With this proposal, we invite abstracts on one of the panel topics.
Oral presentations will be 20 minutes long, followed by 10 minutes for discussion. Please indicate in your submission that your abstract should be considered for the panel. For the panel we only accept oral presentations. We will notify the authors by October 31, 2025.

Submission guidelines: Abstracts must be no longer than 500 words, including references, font size 12, single-spaced. Submissions are limited to a maximum of one individual and one joint abstract per author (or two joint abstracts) for the entire conference. Submissions will open September 1, 2025. Please submit here: https://www.sel.edu.es/liv-simposio-madrid-2026/.

Abstracts should use the corresponding template (“plantilla para el envío de propuestas”) and be submitted in English or Spanish via the registration website (“página web para la solicitud de participación”) (https://www.sel.edu.es/liv-simposio-madrid-2026/cuotas-e-inscripcion/).

Organizing Committee of the Panel:
Alejo Alcaraz, Universidad Autónoma de Madrid (UAM)
[email protected]
Adolfo Ausín, Michigan State University (MSU)
[email protected]
José Camacho, University of Illinois Chicago (UIC)
[email protected]
Carlos Martínez-García, Universidad Complutense de Madrid (UCM)
[email protected]