All German modal particles share important common properties. However, in a diachronic perspective, their origin has often been explained by assuming that they have grammaticalized from different types of lexemes belonging to several word classes: adjectives, adverbs, conjunctions, etc. The present paper intends to dissipate some erroneous assumptions about the grammaticalization paths of modal particles and to offer a novel syntactic approach that explains their origin and development. By following and elaborating on some recent ideas, I will explore the hypothesis that modal particles have an adverbial origin and will provide corresponding evidence. In the syntactic analysis, I will claim that all modal particles originate from specific types of (strong) lower adverb(ial)s that become weak sentential adverbs under reanalysis.
The paper investigates unembedded hwæþer questions in Old English (OE). We argue that they represent an intermediate stage in the development of hwæþer ‘which of the two’ to modern English whether. Syntactically, we find a range of quasi-subordinating uses of hwæþer in questions that all have in common that the speaker expresses a pedagogical question. Pedagogical questions are questions the speaker knows the answer to, but is urging the addressee to consider while drawing their own conclusions. In the OE Boethius, hwæþer can convey this use-conditional pragmatic flavour for polar questions. It thus comes close in function to other use-conditional particles.
We examine the historical reanalysis of the Spanish sequence es que (lit. ‘is that’) as a discourse particle with justificatory and even purely emphatic functions. We argue that the diachronic process involved first the appearance of non-coindexed pro as subject of the copula es ‘is’ and, at a later stage, deletion of the empty pronoun and syntactic restructuring. The restructuring was triggered by the use of the construction in counter-argumentative and other contexts where the semantic content of the null category was not easily recoverable. We also consider parallel developments in Catalan and Portuguese, as well as the borrowing of the particle es que in colloquial Basque.
This study investigates two discourse-related particles, ya and şo, in Laz, an endangered South Caucasian language. We argue that both ya and şo are indexical shift complementizers which can occur without an overt embedding verb, suggesting root complementizer behavior. However, when they appear embedded, the mood specification of the embedding verb determines which of the two will surface, suggestive of complementizer agreement in mood features. We show that, while ya and şo need to be semantically distinct in their root occurrences, there are compositionality challenges against the null hypothesis that ya and şo keep their meanings when embedded. As an alternative to a formal agreement account, we propose to semantically relate the embedded and root occurrences of these complementizers.
This article aims to draw a syntactic analysis that accounts for the differing properties of two sets of discourse particles: Outer (or Final) Particles (OutPs) and Tag Particles (TagPs). In recent years, researchers have built a syntactic model to accommodate OutPs (Haegeman 2014; Wiltschko & Heim 2016; a.o.), but TagPs have received little attention. In order to highlight the different nature of these two sets of particles, I will focus on two segmentally homophonous Basque particles: alaOutP and alaTag . After discussing their distinct prosodic, syntactic and pragmatic properties, I will argue that these particles show structural differences: while OutPs merge in the right periphery of the clause, TagPs are intransitive X0s and head their own Speech Act Phrase (SAP).
On the basis of empirical evidence from various Italian dialects, I argue that primary and secondary interjections lexicalize different functional heads which are computed syntactically at the edge of the clause. Secondary interjections should be clearly distinguished from primary ones; only secondary interjections lexicalizing a SpeechAct° head represent autonomous speech acts and are prosodically and syntactically independent from the co-occurring clause, which they can attract to their specifier position, raising eventually to the adjacent head Speaker° in order to achieve the necessary spatio-temporal contextual anchoring. Primary interjections, which can co-occur with secondary ones and surface clause-initially, lexicalize arguably the highest functional head Speaker°, interacting in interesting ways with lower projections and with the overt realization of the complementizer in Force.
Sentence-final particles (SFPs) in Mandarin Chinese realize the heads of three projections in the rigidly ordered head-final CP ‘Low CP < ForceP < AttitudeP’. Only the highest projection AttitudeP encodes discourse-related properties, whereas ForceP encodes the sentence-type (interrogative, imperative). Low Cs interact with properties of the TP-internal extended verbal projection and are obligatory when acting as (non-default) anchors. They play an important role in determining the temporal interpretation and finiteness in Mandarin Chinese and can therefore no longer be neglected by studies addressing these issues. There is no evidence for an “incremental” acquisition “up the tree” of the different projections in the split CP nor for the acquisition of TP prior to CP, as postulated by the cartographic approach.
In this work we study the meaning and use of the Basque particle bide. We contend that, uttering a bide-sentence, the speaker asserts the proposition she would assert had she uttered a sentence without bide, but conveys additional evidential and doxastic information. She conveys that the evidence for the belief she expresses is indirect and that she is not absolutely certain on its truth–although it seems that the latter is sometimes cancellable. To put it in speech-act theoretic terms, bide is an illocutionary force indicator with no contribution to the propositional content of the speech act, which imposes to the assertion a certain preparatory condition and, perhaps, a constraint on the degree of strength of the belief expressed.
This work attempts to reduce the properties of three German discourse particles (DPs), ja, nicht and etwa, to the basic building blocks of a formal discourse model (Farkas & Bruce 2010). We propose a definition of DPs as speech act modifiers that restricts the space of allowed variation of their meanings, arguing against previous approaches in terms of speaker attitudes. Speech acts modified by ja update the Common Ground (CG) directly; previous characterizations of the epistemic status of the proposition arise as descriptions of common justifications for such an imposed CG update. Etwa and nicht turn open polar questions with two default resolutions into questions with only one unmarked resolution; epistemic or bouletic attitudes arise as frequent connotations.
All German modal particles share important common properties. However, in a diachronic perspective, their origin has often been explained by assuming that they have grammaticalized from different types of lexemes belonging to several word classes: adjectives, adverbs, conjunctions, etc. The present paper intends to dissipate some erroneous assumptions about the grammaticalization paths of modal particles and to offer a novel syntactic approach that explains their origin and development. By following and elaborating on some recent ideas, I will explore the hypothesis that modal particles have an adverbial origin and will provide corresponding evidence. In the syntactic analysis, I will claim that all modal particles originate from specific types of (strong) lower adverb(ial)s that become weak sentential adverbs under reanalysis.
The paper investigates unembedded hwæþer questions in Old English (OE). We argue that they represent an intermediate stage in the development of hwæþer ‘which of the two’ to modern English whether. Syntactically, we find a range of quasi-subordinating uses of hwæþer in questions that all have in common that the speaker expresses a pedagogical question. Pedagogical questions are questions the speaker knows the answer to, but is urging the addressee to consider while drawing their own conclusions. In the OE Boethius, hwæþer can convey this use-conditional pragmatic flavour for polar questions. It thus comes close in function to other use-conditional particles.
We examine the historical reanalysis of the Spanish sequence es que (lit. ‘is that’) as a discourse particle with justificatory and even purely emphatic functions. We argue that the diachronic process involved first the appearance of non-coindexed pro as subject of the copula es ‘is’ and, at a later stage, deletion of the empty pronoun and syntactic restructuring. The restructuring was triggered by the use of the construction in counter-argumentative and other contexts where the semantic content of the null category was not easily recoverable. We also consider parallel developments in Catalan and Portuguese, as well as the borrowing of the particle es que in colloquial Basque.
This study investigates two discourse-related particles, ya and şo, in Laz, an endangered South Caucasian language. We argue that both ya and şo are indexical shift complementizers which can occur without an overt embedding verb, suggesting root complementizer behavior. However, when they appear embedded, the mood specification of the embedding verb determines which of the two will surface, suggestive of complementizer agreement in mood features. We show that, while ya and şo need to be semantically distinct in their root occurrences, there are compositionality challenges against the null hypothesis that ya and şo keep their meanings when embedded. As an alternative to a formal agreement account, we propose to semantically relate the embedded and root occurrences of these complementizers.
This article aims to draw a syntactic analysis that accounts for the differing properties of two sets of discourse particles: Outer (or Final) Particles (OutPs) and Tag Particles (TagPs). In recent years, researchers have built a syntactic model to accommodate OutPs (Haegeman 2014; Wiltschko & Heim 2016; a.o.), but TagPs have received little attention. In order to highlight the different nature of these two sets of particles, I will focus on two segmentally homophonous Basque particles: alaOutP and alaTag . After discussing their distinct prosodic, syntactic and pragmatic properties, I will argue that these particles show structural differences: while OutPs merge in the right periphery of the clause, TagPs are intransitive X0s and head their own Speech Act Phrase (SAP).
On the basis of empirical evidence from various Italian dialects, I argue that primary and secondary interjections lexicalize different functional heads which are computed syntactically at the edge of the clause. Secondary interjections should be clearly distinguished from primary ones; only secondary interjections lexicalizing a SpeechAct° head represent autonomous speech acts and are prosodically and syntactically independent from the co-occurring clause, which they can attract to their specifier position, raising eventually to the adjacent head Speaker° in order to achieve the necessary spatio-temporal contextual anchoring. Primary interjections, which can co-occur with secondary ones and surface clause-initially, lexicalize arguably the highest functional head Speaker°, interacting in interesting ways with lower projections and with the overt realization of the complementizer in Force.
Sentence-final particles (SFPs) in Mandarin Chinese realize the heads of three projections in the rigidly ordered head-final CP ‘Low CP < ForceP < AttitudeP’. Only the highest projection AttitudeP encodes discourse-related properties, whereas ForceP encodes the sentence-type (interrogative, imperative). Low Cs interact with properties of the TP-internal extended verbal projection and are obligatory when acting as (non-default) anchors. They play an important role in determining the temporal interpretation and finiteness in Mandarin Chinese and can therefore no longer be neglected by studies addressing these issues. There is no evidence for an “incremental” acquisition “up the tree” of the different projections in the split CP nor for the acquisition of TP prior to CP, as postulated by the cartographic approach.
In this work we study the meaning and use of the Basque particle bide. We contend that, uttering a bide-sentence, the speaker asserts the proposition she would assert had she uttered a sentence without bide, but conveys additional evidential and doxastic information. She conveys that the evidence for the belief she expresses is indirect and that she is not absolutely certain on its truth–although it seems that the latter is sometimes cancellable. To put it in speech-act theoretic terms, bide is an illocutionary force indicator with no contribution to the propositional content of the speech act, which imposes to the assertion a certain preparatory condition and, perhaps, a constraint on the degree of strength of the belief expressed.
This work attempts to reduce the properties of three German discourse particles (DPs), ja, nicht and etwa, to the basic building blocks of a formal discourse model (Farkas & Bruce 2010). We propose a definition of DPs as speech act modifiers that restricts the space of allowed variation of their meanings, arguing against previous approaches in terms of speaker attitudes. Speech acts modified by ja update the Common Ground (CG) directly; previous characterizations of the epistemic status of the proposition arise as descriptions of common justifications for such an imposed CG update. Etwa and nicht turn open polar questions with two default resolutions into questions with only one unmarked resolution; epistemic or bouletic attitudes arise as frequent connotations.