The cartography of discourse particles in Mandarin and Italian

Start - End 
2015 - 2016 (ongoing)



This project compares some aspects of the syntax of Mandarin and Italian. Within the theoretical framework of the cartographic perspective (Cinque 1999, Rizzi 1997, 2001, 2004, Williams 2009, Cinque & Rizzi 2010), I aim at defining a fine-grained hierarchy of discourse particles in Mandarin and Italian. I choose Italian and Mandarin not only because they are my languages of expertise, but also because the comparison between two typologically unrelated languages gives me the opportunity to derive generalizations that have a higher chance of being crosslinguistically valid. This research is specifically based on the Split-CP hypothesis (Rizzi 1997, Benincà 2001, Haegeman 2002, 2003, 2006, Benincà & Poletto 2004), which conceives the CP as a conglomerate of functional projections that, encompassing all discourse-relevant elements, encode information structure via their heads and the specific features they carry. Some of these functional projections represent the codification of the speaker/hearer related particles that codify the speaker’s attitude and his/her commitment towards the content of the utterances and his/her relation towards an interlocutor (Speas & Tenny 2003, Hill 2007a, 2007b, Giorgi 2008, 2009, 2010, Haegeman & Hill 2010, Delfitto & Fiorin 2011, Frascarelli 2012, Haegeman 2012). Discourse particles are main clause phenomena: they are licensed only in those clauses that are endowed with illocutionary force and they interact with the clause type (see Jacobs 1991, Abraham 1991, Zimmermann 2004a, 2004b, Haegeman 2002, 2004a, 2004b, 2006, to appear). Mandarin: The basic assumption is that Mandarin sentence final particles (MSFPs) are heads of functional projections in the CP domain. Several previous studies have investigated the syntactic and semantic properties of MSFPs (Chao 1968, Hu 1981, Li & Thompson 1981, Zhu 1982, Tang 1989, Chu 1998, Wu 2004, Li 2006, Del Gobbo, Munaro & Poletto to appear, Hsieh & Sybesma 2008). In particular Paul (to appear/a, to appear/b) has provided extensive evidence for the existence of a three-layered CP in Chinese:

(1)  AttitudeP > ForceP > low CP > TP

This three-layered CP can account for the ordering restrictions observed for MSFPs:

(2)  Jieshang                 xingren              bu            duo,

      street.up               pedestrian           NEG        many            

      dagai          dou         zai            jiazhong             kan     dianshi       ne   ba/ *ba   ne.

      probably     all           at              home.middle     watch    TV           PRT PRT   PRT PRT


‘There are not many people in the street, probably they are all at home watching TV.’ (Li 2006:101). On the basis of the hierarchy proposed by Paul, I will concentrate on the topmost layer AttituteP, realized by MSFPs such as a, ne, bale etc. More precisely, I would like to examine in more depth the particles in AttitudeP to propose a more fine-grained hierarchy. In a first stage, I will examine the syntactic behavior of MSFPs realizing AttitudeP by analyzing their interaction with adverbs, modals, topicalized and focused elements, and their distribution in different clause types. For instance, in an exclamative, the presence of the particle ba is not acceptable, in contrast to a (cf. (3)).

(3)           a.               Ta            zheme   gao          a/          *ba!

                                 He            so        tall           PRT        PRT

                                ‘How very tall he is!’

My current work on exclamatives (Badan & Cheng 2010a, 2010b, 2012, 2013) focuses on the syntactic, semantic and pragmatic properties of this particle a in the final position of exclamatives. I am also running prosodic experiments to show that the prosodic contour plays a fundamental role in the interpretation of MSFPs (see for instance Shie 1991, Chu 2002, Wu 2004). For the project presented here, I would like to extend the line of research implemented for a to the other MSFPs realizing AttitudeP. As illustrated in (4) (from Zhu 1982: 213), different MSFPs make different contributions to the interpretation of the sentence that they are attached to:


(4) a.       Tāmen yào wǔbǎi  kuài qián ne!   Bù shì ge xiǎo shùmù!

                3PL want  500 CL money   PRT   NEG be CL small sum

                ‘They want (as much as) 500 dollars! That’s not a small sum!’

      b.       Tāmen yào wǔbǎi kuài qián bale! Méi yǒu shénme liǎobùqǐ!

                3PL want 500 CL money PRT NEG have what extraordinary

‘                 They (only) want 500 dollars! That’s nothing extraordinary!’ 

In the second stage, I’ll aim at defining the semantic and pragmatic contribution of MSFPs in different contexts and their contribution to the illocutionary force of the clause. Italian: I intend to compare MSFPs with Italian sentence initial particles (ISIPs) (for instance tanto, almeno, magari, proprio, però, ma, niente, guarda, vedi, senti, dai…) (for discourse particles in Italian see Burkhardt 1985, Held 1985, Bazzanella 1995, 2006, Munaro & Poletto 2002, 2004, Waltereit 2002, Bonvino, Frascarelli & Pietrandrea 2008, Cardinaletti 2011, Coniglio 2008, Coniglio & Zegrean 2012). Unlike MSFPs, ISIPs generally appear at the beginning of the sentence (even if some of them can also appear in sentence final positions) and are considered discourse particles only in certain uses. Such words are, in fact, polyfunctional, i.e. besides being used as particles, they may have other functions, for instance as adverbs, as in the following example:

(5) a.       Tanto non sai cosa c’è là dentro.

               Very not know2SG what there inside

            ‘At any rate/nevertheless, you don't know what is in there’


      b.        Sei tanto bella.

                be2SG very beautiful

               ‘You are very beautiful.’

There are some studies on Italian discourse particles in sentence internal position (for instance Cardinaletti 2011, Coniglio 2008) and in sentence initial positions (Munaro & Poletto 2002, 2004, Waltereit 2002). However, in the literature no explicit cartographic account of ISIPs has been provided. First, I intend to give a broad classification of ISIPs in Italian, identifying their etymology and conducting a brief diachronic investigation. Then, I aim at testing the behavior of ISIPs: whether they can co-occur with other particles, their ordering restrictions (see for instance the co-occurrence of ma ‘but’ and dai ‘give’ (4)), their interaction with adverbs, modals, and topicalized and focused phrases (see for instance the restrictions of the co-occurrence of niente ‘nothing’ and the topicalized object Giovanni in (5)). I want to analyze their semantic and pragmatic contribution in different contexts and sentence types, and define their contributions to illocutionary force. Determining the semantic contribution of the particles to illocutionary force might be crucial to explain their root status, since speech acts are no longer a prerogative of root clauses, (see Krifka’s (2013) analysis of embedded speech acts); in this way, the reason why all discourse particles of a certain class are a root phenomenon becomes even more interesting, and significantly interacts with Odysseus project on root phenomena. Importantly, I also intend to analyze the prosodic integration of ISIPs with the rest of the sentence, a factor that seems to play a crucial role in their interpretation.


(6)           Ma dai/ *dai ma andiamo!

               PRT PRT PRT PRT go1PL

              ‘Let’s go!’


(7)           Niente, Giovanni/*Giovanni niente non l’ ho più rivisto.

                PRT Giovanni        Giovanni PRT   NEG CL have1SG anymore see again

               ‘Giovanni, I haven’t seen him anymore.’

Research questions: The aim of this research program is to provide a fine-grained cartography of discourse particles in Mandarin and in Italian. Comparatively, I wish to understand whether we can find similarities between the cartography of MSFPs and ISIPs that support a cross-linguistic account of discourse related projections in the sentence periphery. In particular, following Haegeman’s feature sharing approach, I intend to determine what the shared features in the speaker/hearer field are and to understand whether the speaker/hearer related projections can be unified in one unique structural ‘field’ or ‘zone’ (Haegeman 2010a, Haegeman and Hill 2012, to appear, Haegeman to appear, see Chomsky 2001 for the concept ‘phase’). I also intend to analyze the internal structure of the speaker/hearer field and determine whether this field can be further divided in two sub-fields: one exclusively speaker-related and one hearer-related. Finally, I wish to address the question of how the speaker/hearer field interacts with other structural CP fields.