• Nu S-Au Găsit Rezultate

BETWEEN TENSE AND Φ. DEFAULT MARKING IN SOUTHERN ITALIAN DIALECTS

N/A
N/A
Protected

Academic year: 2022

Share "BETWEEN TENSE AND Φ. DEFAULT MARKING IN SOUTHERN ITALIAN DIALECTS "

Copied!
15
0
0

Text complet

(1)

BETWEEN TENSE AND Φ. DEFAULT MARKING IN SOUTHERN ITALIAN DIALECTS

GIUSEPPE TORCOLACCI1

Abstract. This paper investigates the spell-out of a subset of φ features, i.e.

Person and Number, encoded on present perfect and pluperfect auxiliaries in a subset of Southern Italian dialects. We will claim that the overt marking of Person and Number on present perfect and pluperfect auxiliaries in these dialects depends on the application of a post-syntactic operation called Default Marking, according to which a φ feature gets overtly expressed at PF only if its degree of markedness is uniform with that expressed by Tense.

Keywords: auxiliaries, markedness, default, Raddoppiamento Fonosintattico, metaphony.

1. INTRODUCTION

From a cross-linguistic study, Forchheimer (1953: 6) observes that languages tend to exhibit a mismatch in the overt marking of person on finite verbs. While 1st and 2nd person are generally marked on finite verbs, 3rd person is not, as shown by the singular paradigms of the present perfect and pluperfect auxiliaries in (1a) and (1b), respectively2.

(1) Italian a.3

ho mangiato/ dormito H.pres.1sg eaten/slept ‘I have eaten/slept’

hai mangiato/ dormito H.pres.2sg eaten/slept ‘you have eaten/slept’

ha mangiato/ dormito H.pres.3sg eaten/slept ‘(s)he/it has eaten/slept’

abbiamo mangiato/ dormito H.pres.1pl eaten/slept ‘we have eaten/slept’

avete mangiato/ dormito H.pres.2pl eaten/slept ‘you have eaten/slept’

hanno mangiato/ dormito H.pres.3pl eaten/slept ‘they have eaten/slept’

      

1 University of Leiden, [email protected]. 

2 In Italian, the selection of perfect auxiliaries in the active voice is determined by the semantico-syntactic properties, i.e. Aktionsart, of the past participle the auxiliary merges with (cf. Perlmutter 1978, Burzio 1986, Hubert and Rindler-Schjerve 1987, Chierchia 1989, Legendre 1989, Van Valin 1990, Loporcaro 1988 and Sorace 2000, a.o.). While HAVE is selected by accusative and unergative past participles, BE is selected by unaccusative past participles. The same situation is attested for French, within Romance, and German, Dutch and Danish, within Germanic.  

3 In this paper, the gloss H refers to the morphophonological realization of HAVE.

RRL, LX, 2–3, p. 173–187, Bucureşti, 2015

(2)

  b.

avevo mangiato/ dormito H.past.1sg eaten/slept ‘I had eaten/slept’

avevi mangiato/ dormito H.past.2sg eaten/slept ‘you had eaten/slept’

aveva mangiato/ dormito H.past.3sg eaten/slept ‘(s)he/it had eaten/slept’

avevamo mangiato/ dormito H.past.1pl eaten/slept ‘we had eaten/slept’

avevate mangiato/ dormito H.past.2pl eaten/slept ‘you had eaten/slept’

avevano mangiato/ dormito H.past.3pl eaten/slept ‘they had eaten/slept’

The singular paradigms in (1) show that 1st and 2nd person are always marked by means of a dedicated φ marker. 3rd person, on the contrary, is never marked. The 3rd singular auxiliaries in (1a) and (1b) admit the overt realization of the morpheme [a] in word-final position. This morpheme also appears in the plural paradigms and is followed by another morpheme expressing φ information4.

The pattern of φ marking observed for Italian in (1) is unattested for a group of Southern Italian dialects, namely for a group of dialects spoken in the geolinguistic area that stretches from central Apulia to Salento, on the Adriatic side, and from central Campania to northern Calabria, on the Tyrrhenian side. In these dialects, present perfect auxiliaries generally disallow the overt marking of 2nd person in the singular paradigm5. The absence of a φ marker realized in word-final position of 2nd singular present perfect HAVE leads to the rise of a syncretism between 2nd and 3rd singular present perfect forms.

In spite of this syncretism, 3rd singular present perfect HAVE, differently from 2nd singular present perfect HAVE, is followed by a past participle exhibiting a double consonant in word-initial position. Double consonants in initial positions in Southern Italian dialects are considered to be instances of Raddoppiamento Fonosintattico (henceforth RF)6. These facts are represented in (2). In this paper, we will refer only to the dialect of Mola di Bari, spoken south of Bari. All other Southern Italian dialects featuring the same φ marking strategies observed for the dialect of Mola di Bari will not be presented here. For a well documented list of dialects patterning in the same way as Mola di Bari, as far as the marking of φ on present perfect and pluperfect auxiliaries is concerned, see Torcolacci (2015).

      

4 We postulate that the vowel [a] preceding the φ markings in (1a) and (1b) is a morpheme that does not express person information. This assumption is based on the empirical evidence that [a] is spread to all forms in the paradigms, thus being not sensitive to the type of person feature encoded on the auxiliaries.  

5 Dialects belonging to this geolinguistic area normally select HAVE as a present perfect auxiliary combining with all types of past participles. This pattern is also observed for Spanish, Romanian and Extreme Southern Italian dialects (cf. Manzini and Savoia 2005), within Romance. In the pluperfect, however, a handful number of dialects spoken in and around the Murge and central Campania choose BE, instead of HAVE, for all persons in the paradigm. The choice of BE as a pluperfect auxiliary is insensitive to the nature of the past participle the auxiliary combines with. For a survey of these data, see Manzini and Savoia (2005) and Cennamo (2010).  

6 The literature of RF, as well as the interaction between RF and 3rd present perfect auxiliary HAVE, will be highlighted in §2. 

(3)

(2) Mola di Bari (Apulo Barese) – present perfect construction

aɟɟ 'fatt/ par'lə:t/ par'tʉ:t H.pres.1sg done/spoken/left ‘I have done/spoken/left’

a 'fatt/ par'lə:t/ par'tʉ:t H.pres.2sg done/spoken/left ‘you have done/spoken/left’

a f'fatt/ ppar'lə:t/ ppar'tʉ:t H.pres.3sg done/spoken/left ‘(s)he/it has done/spoken/left’

am 'fatt/ par'lə:t/ par'tʉ:t H.pres.1pl done/spoken/left ‘we have done/spoken/slept’

avet 'fatt/ par'lə:t/ par'tʉ:t H.pres.2pl done/spoken/left ‘you have done/spoken/slept’

an 'fatt/ par'lə:t/ par'tʉ:t H.pres.3pl done/spoken/left ‘they have done/spoken/slept’

On the other hand, pluperfect auxiliaries of the dialect of Mola di Bari, as well as of all the other dialects spoken in the same geolinguistic area, allow the overt marking of 2nd person, both in the singular and in the plural paradigm. The marking of 2nd person operates by the selection of the vowel [i] in stressed position. In the singular paradigm, 1st and 3rd person pluperfect auxiliaries are not marked for their φ features and are thus represented by means of a syncretic exponent featuring the vowel [ɒ] in stressed position. We will consider the presence of the high-vowel [i] on 2nd singular (and plural) pluperfect HAVE as an instance of metaphony7. These facts are illustrated in (3).

(3) Mola di Bari (Apulo-Barese) – pluperfect construction

a'vɒ:v 'fatt/ par'lə:t/ par'tʉ:t H.past.1sg done/spoken/left ‘I had done/spoken/left’

a'vi:v 'fatt/ par'lə:t/ par'tʉ:t H.past.2sg done/spoken/left ‘you had done/spoken/left’

a'vɒ:v 'fatt/ par'lə:t/ par'tʉ:t H.past.3sg done/spoken/left ‘(s)he/it had done/spoken/left’

a'vɛmm 'fatt/ par'lə:t/ par'tʉ:t H.past.1pl done/spoken/left ‘we had done/spoken/left’

avi:vər 'fatt/ par'lə:t/ par'tʉ:t H.past.2pl done/spoken/left ‘you had done/spoken/left’

avɛ:vən 'fatt/ par'lə:t/ par'tʉ:t H.past.3pl done/spoken/left ‘they had done/spoken/left’

The paradigms in (2) and (3) show that:

− 2nd and 3rd singular present perfect HAVE are syncretic;

− 1st and 3rd singular pluperfect HAVE are syncretic;

− in the present perfect, only 3rd singular present perfect HAVE is followed by RF;

− in the pluperfect, only 2nd singular (and plural) pluperfect HAVE is/are affected by metaphony.

Based on these empirical observations, the gist of this paper is to investigate whether the different types of syncretism observed in (2) and (3), as well as the application of RF and metaphony occurring within a well defined set of auxiliaries in these paradigms, are to be considered ‘accidental’ or rather dependent on the application of a specific morphological rule. In this paper, it will be proposed that the overt marking of φ features in the paradigms in (2) and (3) is driven by the application of a post-syntactic operation called Default Marking, whereby φ features encoded on present perfect and pluperfect auxiliaries       

7 The literature of metaphony, as well as the interaction between metaphony and 2nd singular pluperfect HAVE, will be analyzed in §2. 

(4)

 

get overtly expressed at PF only if their degree of markedness coincides with that expressed by the feature Tense. The framework we will refer to is Distributed Morphology (cf. Halle and Marantz 1993, 1994, Calabrese 1994, Harley 1994, Harris 1994, Embick 1995, Noyer 1997, Harley and Noyer 1999, a.o.), which predicts that morphology is an independent component of the grammar sandwiched between syntax and phonology. In the present paper, the Default Marking operation will be considered to apply in the morphological component.

This paper is structured as follows: in §2, the phenomenon of RF attested after 3rd singular present perfect southern Italian auxiliaries and metaphony found on 2nd singular pluperfect southern Italian auxiliaries will be discussed. §3 will consider the post-syntactic operation of Default Marking. §4 summarizes and concludes the paper.

2. RADDOPPIAMENTO FONOSINTATTICO AND METAPHONY ON PERFECT AUXILIARIES

According to Schuchardt (1874), Hall (1964), Loporcaro (1997b) and Waltereit (2004), the presence of RF in Southern Italian dialects strictly depends on the application of the phonological process of Regressive Consonant Assimilation, or RCA, that took place in the period of transition from Latin to Southern-Italo Romance. More precisely, Schuchardt, Hall, Loporcaro and Waltereit claim that word-final consonants in Latin got assimilated to the first consonant of the next word in diachrony. This process is taken to have given rise to RF8. These facts are shown in (4).

(4) 3rd singular present perfect HAVE

a. Latin: *HA(BE)T + CVCV b. Southern Italian dialects: /a/ + CVCV

σ9 σ σ σ

μ μ μ μ μ μ μ μ

a t C V a C V

In (4a), the word *HA(BE)T ends in a consonant, namely /t/. This consonant became assimilated to the first consonant of the next word in the development from Latin to Southern Italian dialects and, as a result, a geminate was realized. If we consider the presence of RF triggered by 3rd singular present perfect HAVE in (2) and (4) as deriving       

8 Loporcaro (2007) argues that RF can be of two types: regular and irregular RF. While irregular RF results from the encounter of two consonants at word-boundaries, regular RF is induced by stress on oxytonic words (cf. Standard Italian: fà bbène ‘(s)he/it does it well’ versus fài bène ‘you do it well’). Irregular RF is not found only in Southern Italian dialects, but also in Central Italian dialects and Standard Italian. Regular RF, on the other hand, is found only in Central Italian dialects and Standard Italian, with the exclusion of the southern dialects.  

RF

(5)

from the application of RCA in diachrony, then we would not understand why RF is not observed for 2nd singular present perfect HAVE. 2nd singular present perfect HAVE in the dialect of Mola di Bari in (2), as well as in all Southern Italian dialects, never admits RF.

These facts are sketched below in (5).

(5) 2nd singular present perfect HAVE

a. Latin: *HA(BE)S + CVCV b. Southern Italian dialects: /a/ + CVCV

σ σ σ σ

μ μ μ μ μ μ μ μ

a s C V a C V

In (5b), RF is not produced. This might due to the fact that the consonant /s/ in word- final position of *HA(BE)S did not undergo the RCA process in diachrony. The reason why RCA was not active in (5) might depend on the fact that the mora associated to /s/ got deleted in the diachronic evolution from Latin to southern Italo-Romance. On the other hand, the mora associated with /t/ of *HA(BE)T was not deleted, and for this reason the consonant /t/ was assimilated to the first one of the next word. Because /t/ of *HA(BE)T is the Latin segment that expresses 3rd singular, then we propose that the first consonant of the geminate generated through RF in (4b) also expresses 3rd singular. The difference between /t/ of *HA(BE)T in (4a) and RF in (4b) consists in the fact that while 3rd singular in (4a) is expressed by /t/, 3rd singular is expressed by the double consonant in (4b), namely by the onset of the first syllable of the next word that undergoes regressive spreading.

If we now turn to the singular paradigm of the pluperfect construction in (3), we observe that metaphony is found only when the auxiliary is in the 2nd singular, and never when this expresses 1st or 3rd singular. In the traditional literature (cf. Maiden 1991, Calabrese 1998, 1999), metaphony is considered to be a phonological process whereby a stressed vowel is raised when the following syllable(s) also contain(s) a high vowel. If we consider metaphony on 2nd singular pluperfect HAVE in (3) as driven by the presence of a high vowel in final position, then we would not understand why 2nd singular present perfect HAVE does not feature metaphony in those dialects that host a high vowel in word-final position. Indeed, 2nd singular present perfect HAVE in a group of Salentino and Lucanian dialects is endowed with a high vowel in final position, the presence of which does not trigger the heightening of the stressed vowel (Maglie and Giurdignano/Uggiano La Chiesa [Southern Salentino]: ai ca'matu ‘you.sg have called’, cf. Manzini and Savoia 2005 II:

753−754). This suggests that the presence of a high vowel in word-final position does not a priori triggers metaphony on perfect auxiliaries. If this were true, then we would expect to find metaphony in all those 2nd singular present perfect auxiliaries that host the vowel [i] in word-final position.

If the high vowel [i] in word-final position of 2nd present perfect HAVE of the dialects of Maglie and Giurdignano/Uggiano la Chiesa discussed above is considered to be the morphophonological expression of a morpheme encoding 2nd person, then we can argue

no RF

(6)

 

that metaphony is also a way of expressing 2nd person on a pluperfect auxiliary. The difference between the marking of 2nd singular by means of /i/ in word-final position and through metaphony in word-internal position would then lie on the locus where this φ value gets morphophonologically expressed.

The facts presented in this section have shown that RF and metaphony are phonological processes that signal the presence of a given type of Person feature encoded on present perfect and pluperfect auxiliaries. The next section will consider why the dialect of Mola di Bari in (2) and (3), together with other dialects spoken in the same geolinguistic area, tend to overtly mark 1st and 3rd person on present perfect auxiliaries and 2nd person on pluperfect auxiliaries.

3. THE POST-SYNTACTIC OPERATION OF DEFAULT MARKING 3.1. The inventory of φ features on perfect auxiliaries and their acquisition Let us reproduce in (6) and (7) the present perfect and pluperfect paradigms of the dialect of Mola di Bari presented in (2) and (3), respectively.

(6)

aɟɟ 'fatt/ par'lə:t/ par'tʉ:t H.pres.1sg done/spoken/left ‘I have done/spoken/left’

a 'fatt/ par'lə:t/ par'tʉ:t H.pres.2sg done/spoken/left ‘you have done/spoken/left’

a f'fatt/ ppar'lə:t/ ppar'tʉ:t H.pres.3sg done/spoken/left ‘(s)he/it has done/spoken/left’

am 'fatt/ par'lə:t/ par'tʉ:t H.pres.1pl done/spoken/left ‘we have done/spoken/left’

avet 'fatt/ par'lə:t/ par'tʉ:t H.pres.2pl done/spoken/left ‘you have done/spoken/left’

an 'fatt/ par'lə:t/ par'tʉ:t H.pres.3pl done/spoken/left ‘they have done/spoken/left’

(7)

a'vɒ:v 'fatt/ par'lə:t/ par'tʉ:t H.past.1sg done/spoken/left ‘I had done/spoken/left’

a'vi:v 'fatt/ par'lə:t/ par'tʉ:t H.past.2sg done/spoken/left ‘you had done/spoken/left’

a'vɒ:v 'fatt/ par'lə:t/ par'tʉ:t H.past.3sg done/spoken/left ‘(s)he/it had done/spoken/left’

a'vɛmm 'fatt/ par'lə:t/ par'tʉ:t H.past.1pl done/spoken/left ‘we had done/spoken/left’

avi:vər 'fatt/ par'lə:t/ par'tʉ:t H.past.2pl done/spoken/left ‘you had done/spoken/left’

avɛ:vən 'fatt/ par'lə:t/ par'tʉ:t H.past.3pl done/spoken/left ‘they had done/spoken/left’

The overt marking of φ features in the singular paradigms in (6) and (7) is in complementary distribution. While 1st and 3rd person are overtly marked in the singular paradigm in (6), (7) admits the only overt marking of 2nd person in the singular9.

      

9 We will not take into account the morphophonological markedness of Person and Number feature in the plural paradigms in (6) and (7). For an overview of the morphophonological

(7)

At this point, it is relevant to consider the type of Person and Number feature that gets overtly expressed in the singular paradigms in (6) and (7). Harley and Ritter (2002) propose that φ features encoded on DP pronouns are structurally organized within a geometry, whose hierarchy is shown in (8).

(8)

The geometry in (8) shows that morphosyntactic φ features are in a dependency relation. The nodes Speaker and Addressee, for instance, branch below Participant, while the nodes Minimal, Group and Class are the dependents of Individuation. According to Harley and Ritter (2002), Participant and Individuation express person and number properties, respectively. On the other hand, Speaker and Addressee express 1st and 2nd person, respectively, while Minimal and Group express singular and plural, respectively.

Class corresponds to the mother node expressing gender properties. In line with the geometry in (8), we assume that 3rd person pronouns completely lack a person feature. This is due to the fact that the Person, or Participant, feature is encoded only on 1st and 2nd person pronouns, namely on those pronouns that encode Speaker and Addressee10. The assumption that only 1st and 2nd person pronouns are endowed with a Person feature, with the exclusion of 3rd person ones, goes back to Forchheimer (1953) and Benveniste (1966, 1971). Forchheimer, for instance, argues that the dichotomy between 1st and 2nd pronouns, on one hand, and 3rd pronouns, on the other, lies on the fact that “whoever does not act a rôle in the conversation either as speaker or as addressee remains in the great pool of impersonal, referred to as ‘third person’” (Forchheimer 1953: 5–6)11. This said, the difference between 1st, 2nd and 3rd pronouns consists in the fact that only the first two        

markedness of Person and Number feature in the plural paradigm of present perfect and pluperfect constructions in the dialect of Mola di Bari, and neighboring dialects, see Torcolacci (2015). 

10 Béjar and Řezáč (2009) propose an alternative idea as far as the make-up of features contained in 3rd person pronouns is concerned. They assume that 3rd person pronouns are not all the same, and that 3rd person animates only contain a Person feature, with the exclusion of Participant and Author.  

11 Other researchers have claimed that only 1st and 2nd person pronouns, with the exclusion of 3rd person ones, are endowed with a Person, or Participant feature. For a survey, see Wundt (1911), Schimdt (1919), Jespersen (1924), Bloomfield (1933), Buehler (1934), Jackobson (1971).  

(8)

 

encode a Participant feature, while the latter does not, and expresses Individuation. These facts are made explicit in (9).

(9)

1st person DP pron. 2nd person DP pron. 3rd person DP pron.

Participant, Speaker Participant, Addressee Individuation, Minimal

The geometry in (8), as well as the table in (9), indicate that Speaker and Minimal, differently from Addressee, are curly underlined. The use of this diacritic signals that these two features are those that are acquired early in the acquisitional path of pronouns. From a typological study about the acquisition of pronouns, Harley and Ritter (2002) observe that Minimal is the first node acquired within the Individuation domain, while Speaker is the first node acquired within the Participant domain. Since Speaker and Minimal are the first nodes acquired, these are considered as defaults. This means that before the acquisition of Addressee, Participant bears the default interpretation of Speaker. As for Individuation, this bears the default interpretation of Minimal before Group is acquired (cf. (10) and (11)).

(10) a. Stage 1 (early acquisition of Speaker) b. Stage 2 (acquisition of Addressee) Participant = Speaker Participant

Speaker Addressee

(11) a. Stage 1 (early acquisition of Minimal) b. Stage 2 (acquisition of Group) Individuation = Minimal Individuation

Minimal Group The early acquisition of Speaker and Minimal, as opposed to Addressee and Group, is also observed for agreement markers. Ackema and Neeleman (2012) show in fact that 1st and 3rd persons agreement markers are always acquired before 2nd person agreement markers. For this reason, we expect 1st and 3rd agreement markers to follow the same acquisitional path sketched for DP-pronouns in (10) and (11). In addition, we propose that 1st, 2nd and 3rd person agreement markers are also endowed with the features Speaker, Addressee and Minimal, respectively. This depends on the fact that after the Agree relation between a DP-subject and a verb is instantiated, the φ values contained on the DP-subject are copied onto the verb (Chomsky 1995, 2000, 2001). After Agree takes place, the φ features of the DP-subject are also valued on the verb.

3.2. Perfect auxiliaries and the Default Marking

In this subsection, we argue that perfect auxiliaries in Southern Italian dialects, similarly to what has been assumed for lexical verbs in other Romance languages, are

(9)

merged in Infl°. The past participle, on the other hand, is merged in an aspectual head that linearly follows the auxiliary12. This syntactic configuration is sketched in (12).

(12)

Chomsky (1981) postulates the presence of an abstract category called INFL where Tense and subject-verb agreement features are encoded. The same idea has been adopted by Ritter and Wiltschko (2010), according to whom, building on Ritter and Wiltschko (2009), Tense and Person features are encoded within the universal head/position INFL. Together with Tense and Person, INFL can encode the feature Location (cf. Ritter and Wiltschko (2010)). The syntactic configuration of INFL is given in (13), where the category INFL is replaced by Infl°, which in the present treatment corresponds to a syntactic head. The feature Person, on the other hand, is replaced by φ.

(13) Infl°

Tense φ Location

Following Ritter and Wiltschko (2010), we consider the value of the feature Tense in Infl° to be dependent on a relation between the event and the utterance time of the sentence.

While Present is expressed in the case the event and the utterance time of the sentence       

12 We will not consider here what the exact merging site of the aspectual head is that contains the past participle of the constructions in (6) and (7). In the recent literature, it has been claimed that two aspectual heads are present in syntax: one corresponds to a head that encodes inner aspect, while the other corresponds to a syntactic head that expresses outer aspect (cf. Slabakova 2001, Ramchand 2008, Travis 2010). Inner aspect is concerned with inherent boundaries of events and the telic/atelic distinctions, while outer aspect is concerned with actual boundaries of events. As Slabakova (2001), Ramchand (2008) and Travis (2010) point out, outer aspect is merged outside the vP layer, while inner aspect is merged inside the vP layer. 

(10)

 

coincide, Past is conveyed when the event and the utterance time of the sentence differ in their time reference13. According to Ritter and Wiltschko (2010), the event time is encoded in Spec,VP, while the utterance time is encoded in Spec,InflP. Furthermore, Ritter and Wiltschko (2010) postulate the presence of a [ucoin(cidence)] feature in Infl°, the function of which is to express the relation obtained between the event and the utterance time. If the event and the utterance time are uniform in their reference, then [ucoin] bears a + value. In the opposite situation, namely if the event and the utterance time are not uniform in their reference, then [ucoin] bears a – value. In the present treatment, the [ucoin] feature is encoded in Tense and the event situation is expressed in Spec,AspP. These facts are explained in the representation in (14).

(14)

In the spirit of Holmberg and Roberts (2010), we consider uniformity of feature values to drive unmarked syntactic configurations. According to Holmberg and Roberts, features that express the same values are thought to feed unmarked syntactic configurations.

On the other hand, features that do not express the same values are thought to drive marked syntactic configurations. With reference to the EPP feature, for instance, Holmberg and Roberts (2010) claim that if all syntactic heads endowed with an EPP feature are endowed with a +/-EPP feature, then an unmarked syntactic configuration is obtained. Conversely, if some syntactic heads are endowed with a +EPP feature, while some other express a –EPP feature, then a marked syntactic configuration is obtained14. Based on this assumption, we claim that the [ucoin] feature in Tense gets an unmarked, or default, specification only if the event and the utterance time share the same time reference. This is to say that [+coin] is a default because the event and the utterance situation refer to the same unit of time, thus       

13 A similar idea was put forward by Comrie (1985). He assumes that Present is expressed when the moment of speech is simultaneous with the event time, while Past is conveyed when the event time precedes the moment of speech. 

14 “In these terms, rigidly head-final languages are relatively unmarked, as of course are rigidly head-initial languages, while ‘mixed’ languages are relatively ‘marked’ (and one can in principle quantify exactly how marked different types of mixed systems would be)” (cf. Holmberg and Roberts 2010: 40).  

(11)

being uniform in their time reference. On the other hand, the feature [-coin] is marked since it signals that the event and the utterance time are not uniform in their reference (cf. (15)).

(15)

Based on these facts, we propose that the value expressed by [ucoin] in the periphrastic constructions in (6) and (7) is able to determine the set of φ features encoded on a group of southern Italian present perfect auxiliaries that get overtly spelled-out at PF.

More precisely, we postulate the application of a post-syntactic operation called Default Marking, whose definition is given in (16).

(16) Default Marking

The morphological marking of a φ feature can only take place if all features bear the same markedness on the functional head that hosts them15.

What (16) says is that a φ feature can be overtly expressed, thus being phonologically realized, only if its degree of markedness is uniform with other features encoded in the same functional head. Thus, the definition in (16) predicts that a default φ feature encoded on a perfect auxiliary can be overtly expressed only if [ucoin] expresses a + value. On the other hand, a marked φ feature encoded on a perfect auxiliary can be overtly expressed only if [ucoin] encodes a – value. These facts are shown in (17) and (18).

In (17), [ucoin] is valued for +, thus for a default value, and for this reason only Speaker and Minimal get overtly marked at PF. This depends on the fact that Speaker and Minimal are defaults for Participant and Individuation, respectively, and for this reason they can be overtly marked because they share the same degree of markedness with [+coin]. The overt marking of Speaker results through the selection of the inflectional marker /ɟɟ/ in word-final position, while the marking of Minimal results by means of RF.

Addressee, which is the marked feature in the domain of Participant, does not get marked.

This depends on the fact that Addressee and [+coin] do not bear the same degree of markedness, and for this reason the Default Marking operation sketched in (16) cannot be triggered post-syntactically.

      

15 The use of Default Marking, as a term, is justified by the fact that a default morphosyntactic configuration is attested (only) when two morphosyntactic features hosted on the same syntactic head are uniform in their values. In a default context, the overt marking of a feature (a φ feature in our case) can apply.  

(12)

 

(17)

In (18), [ucoin] is valued for -, which is a marked value. In this context, only Addressee, with the exclusion of Speaker and Minimal, get overtly marked at PF. This depends on the fact that Addressee and [-coin] express the same degree of markedness, and for this reason Addressee is marked at PF. The overt marking of Addressee results by means of metaphony, thus by the selection of /i/ in stressed position. Speaker and Minimal, in being the default features within the Participant and Individuation domains, respectively, do not get phonologically marked and, as a result, the two auxiliaries encoding these two features select a syncretic exponent.

The application of the post-syntactic operation Default Marking that we have treated in this section is now able to capture the complementary distribution of morphological markedness of φ features in the present perfect and pluperfect paradigms in (2)−(6) and (3)−(7).

Torcolacci (2015) shows that the dialect of Mola di Bari and surrounding dialects allow Default Marking also in the case of lexical verbs, determiners and demonstratives.

This means that Default Marking does not only affect auxiliaries, but also other syntactic elements. As for the geolinguistic extension of this phenomenon, Torcolacci observes that Default Marking applies only within a set of dialects spoken in Southern Italy: not all southern Italian dialects are affected by this post-syntactic phenomenon. In this respect, one would postulate Default Marking to be a morphological parameter that is found only for some languages.

4. CONCLUSIONS

In this paper, we have considered the morphological markedness of φ features that occur on present perfect and pluperfect auxiliaries in a number of southern Italian dialects.

(13)

In a group of Italian dialects spoken between central/southern Calabria-central Apulia and northern Calabria, present perfect and pluperfect auxiliaries do not show the same type of morphological markedness, as far as φ features are concerned. While present perfect auxiliaries in the singular paradigm allow the overt marking of 1st and 3rd person only, with the exclusion of 2nd person, pluperfect auxiliaries in the singular paradigm admit the overt marking of 2nd person only. In present perfect auxiliaries, the overt marking of 1st person results through the selection of an independent φ marker, while 3rd person is marked through RF. In pluperfect auxiliaries, on the other hand, 2nd person is marked through metaphony. With reference to Harley and Ritter (2002), we have considered 1st, 2nd and 3rd person singular as instances of the features Speaker, Addressee and Minimal. Speaker and Minimal, differently from Addressee, are considered to be default features, while Addressee is marked. This depends on the fact that Speaker and Minimal are acquired early in the acquisition path, while Addressee is acquired late.

In §3, we have considered perfect auxiliaries in southern Italian dialects as syntactic heads merged in Infl°. Infl°, according to Ritter and Wiltschko (2010), is a syntactic head that expresses three deictic categories, which include Person (or φ), Tense and Location.

Based on Ritter and Wiltschko (2010), we have postulated the presence of an [ucoin]

feature in Infl°, more precisely in Tense, whose function is to overtly express the relation between the event and the utterance time. If the event time coincides with the utterance time, then [ucoin] is valued as +, which is default, while if the event time and the utterance time do not coincide, then [ucoin] is valued as -, which is marked. In the spirit of Holmberg and Roberts (2010), we have considered uniformity of feature values to drive unmarked syntactic configurations. Based on this assumption, we have considered the presence of two features on the same functional head that express the same degree of markedness to license unmarked configurations. In our account, uniformity of feature values on a functional head triggers the application of Default Marking, that is supposed to apply in morphology.

Default Marking predicts that φ features can be overtly marked only if their degree of markedness is the same as that of [ucoin]. If [ucoin] bears a + value, which is default, then only default φ features are overtly expressed. On the contrary, if [ucoin] bears a – value, which is marked, then only marked φ features are overtly expressed.

Given these facts, one should consider whether Default Marking should be treated as a purely morphological operation. One motivation for assuming that Default Marking applies in morphology might depend on the fact that it is activated only after [ucoin] in Tense and φ features in Infl° have been valued. Since valuation of features applies in syntax, there are obvious reasons to assume that Default Marking applies after it, namely in the morphological component. If Default Marking is a parameter, then we should consider whether its domain of application might include any language or, conversely, if some languages might not be affected by it. A tentative answer is to consider Default Marking as an operation that can apply only in languages with fusional morphology; agglutivative languages, indeed, allow the overt marking of every morpheme encoded on every syntactic head. Fusional languages, on the contrary, do not allow the overt marking of all morphemes since the number (and type) of morphemes to be spelled-out at PF must be determined in the morphological component. In this respect, Default Marking might correspond to one of those post-syntactic mechanisms that gets parametrized in a given set of fusional languages.

(14)

 

REFERENCES

Ackema, P., A. Neeleman, 2012, “Person features and syncretism: a case study”. Paper presented at the Syntax Circle, University of Amsterdam.

Béjar, S., M. Řezáč, 2009, “Cyclic Agree”, Linguistic Inquiry, 40, 35−73.

Benveniste, E., 1966, Problèmes de linguistique générale, Paris, Gallimard.

Benveniste, E., 1971, “The nature of pronouns”, in: Problems in general linguistics, University of Miami Press, Coral Gables, 217–222.

Bloomfield, L., 1933, Language, New York, Henry Holt.

Buehler, K., 1934, Sprachtheorie: Die Darstellungsfunktion der Sprache, Jena, Fischer.

Burzio, L., 1986, Italian Syntax, Dordrecht, Reidel.

Calabrese, A., 1994, “Syncretism phenomena in the clitic systems of Italian and Sardinian dialects and the notion of morphological change”, in: J. Beckman (ed.), Proceedings of NELS 25, University of Massachusetts, Amherst, 151−174.

Calabrese, A., 1998, “Metaphony Revisited”, Rivista di Linguistica, 10, 7−68.

Calabrese, A., 2009, “Metaphony”, Ms. University of Connecticut.

Cennamo, M., 2010, “Perfective auxiliaries in the pluperfect in some southern Italian dialects”. in:

R. D’Alessandro, A. Ledgeway, I. Roberts (eds), Syntactic Variation: the Dialects of Italy, Cambridge, Cambridge University Press, 210−224.

Chierchia, G., 1989, “A Semantics for Unaccusatives and its Syntactic consequences”. Ms. Cornell University, Ithaca, N. Y.

Chomsky, N., 1981, Lectures on Government and Binding, Dordrecht, Foris.

Comrie, B., 1985, Tense, Cambridge, Cambridge University Press.

Embick, D., 1995, “Mobile inflections in Polish”, NELS 25, University of Massachusetts, Amherst, 127−142.

Forchheimer, P., 1953, The category of person in language, Berlin, Walter de Gruyter.

Hall, R., 1964, “Initial consonants and syntactic doubling in West Romance”, Language, 40, 551−556.

Halle, M., A. Marantz, 1993, “Distributed Morphology and the Pieces of Inflection”, in: K. Hale, S. Keyser (eds), The View from Building 20, Cambridge, MIT Press, 111−176.

Halle, M., A. Marantz, 1994, “Some key features of Distributed Morphology”, in: A. Carnie, H. Harley (eds), Papers on Phonology and Morphology, MITWPL 21, Cambridge, MIT Press, 275−288.

Harley, H., 1994, “Hug a tree: Deriving the morphosyntactic feature hierarchy”, in: A. Carnie, H. Harley (eds.), Papers on phonology and morphology, MITWPL 21, Cambridge, MIT Press, 275−288.

Harley, H., E. Ritter, 2002, “Person and number in pronouns: a feature-geometric analysis”, Language 78, 482−526 (Page numbers refer to the manuscript).

Harley, H., R. Noyer, 1999, “State-of-the-art article: Distributed morphology”, in: GLOT International 4.

Harris, J., 1994, “The syntax-phonology mapping in Catalan and Spanish clitics”, in: A. Carnie, H. Harley (eds), Papers on phonology and morphology, MITWPL 21. Cambridge, MIT Press, 321−353.

Hayes, B., 1989, “Compensatory Lenghtening in Moraic Phonology”, Linguistic Inquiry, 20, 253−306.

Holmberg, A., I. Roberts, 2010, “Introduction: parameters in minimalist theory’, in: T. Biberauer, A. Holmberg, I. Roberts, M. Sheehan (eds), Null subjects: the structure of parametric variation. Cambridge, Cambridge University Press, 1−57.

Hubert, H., R. Rindler-Schjerve, 1987, “The parameter of auxiliary selection. Italian-German contrasts”, in: Linguistics, 25, 1029−1055.

(15)

Jakobson, R., 1971, “Shifters, verbal categories, and the Russian verb”, in: Selected writings 2, The Hague, Mouton, 130−147.

Jespersen, O., 1924, The Philosophy of Grammar, New York, The Norton Library.

Legendre, G., 1989, “Unaccusativity in French”, in: Lingua, 79, 95−164.

Loporcaro, M., 1988, “History and Geography of Raddoppiamento Fonosintattico: Remarks on the Evoution of a Phonological Rule’, in: P.M. Bertinetto, M. Loporcaro (eds), Certamen phonologicum. Papers from the 1987 Cortona phonology meeting. Torino, Rosenberg/Sellier, 341−387.

Loporcaro, M., 1997, L’origine del raddoppiamento fonosintattico. Saggio di fonologia diacronica romanza, Basel-Tübingen, Francke.

Maiden, M., 1991, Interactive Morphonology: Metaphony in Italy, New York, Routledge.

Manzini, M. R., L. M. Savoia, 2005, I dialetti italiani e romanci. Morfosintassi generativa.

Alessandria, Edizioni dell’Orso.

Noyer, R., 1997, Features, Positions and Affixes in Autonomous Morphological structures. New York, Garland Publishing. Revised version of 1992 MIT Doctoral Dissertation.

Perlmutter, D. M., 1978, “Impersonal Passives and the Unaccusative Hypothesis”, in: Proceedings of the 4th Annual Meeting of the Berkley Linguistic Society, 157−189.

Ramchand, G., 2008, Verb meaning and the Lexicon. A first phase syntax, Cambridge, Cambridge University Press.

Ritter, E., M. Wiltschko, 2009, “Varieties of INFL: TENSE, LOCATION and PERSON”, in: J. van Craenenbroek (ed.), Alternatives to Cartography, Berlin, Mouton de Gruyter, 153−202.

Ritter, E., M. Wiltschko, 2010, “The composition INFL: An exploration of tense, tenseless languages and tenseless constructions”, LingBuzz/001078, July 2010.

Schmidt, P. W., 1919, Die Personalpronomina in den australischen Sprachen, Vienna, Alfred Holder.

Schuchardt, H., 1874, “De quelques modifications de la consonne initiale dans les dialectes de la Sardaigne, du Centre et du Sud de l’Italie”, Romania, 3, 1−30.

Slabakova, R., 2001, Telicity in second language, Amsterdam, John Benjamins.

Sorace, A., 2000, “Gradients in auxiliary selection with intransitive verbs”,Language, 76, 859−890.

Torcolacci, G., 2015, Marking the Default. Auxiliary selection in Southern Italian dialects, Utrecht, LOT.

Travis, L., 2010, Inner aspect: the articulation of VP, Dordrecht, Springer.

Van Valin, R. D., 1990, “Semantic parameters of split intransitivity”. Language, 66, 221−260.

Waltereit, R., 2004, “Fossils of a former phonological rule: Irregular raddoppiamento fonosintattico”, Philologie im Netz 29, 40−59.

Wundt, W, 1911, Die Sprache, Leipzig, Kröner-Engelmann.

Referințe

DOCUMENTE SIMILARE

Keywords: Differential Object Marking, Clitic Doubling, Clitic Left Dislocation, grammaticalization, reanalysis, preposition, (topic) marker, Case, (Old)

The evolution to globalization has been facilitated and amplified by a series of factors: capitals movements arising from the need of covering the external

In 1980, John Naisbitt identified 20 mega trends that have recently appeared in the research environment: transformation of industrial society to informational society, spreading

Using a case study designed for forecasting the educational process in the Petroleum-Gas University, the paper presents the steps that must be followed to realise a Delphi

In our data we have found some instances of object marking errors: in constructions that require obligatory genitive marking the accusative case was used.. This is especially

At the same time, it has been stated that DOM in Estonian is often considered to be the very difficult part of Estonian grammar for L2 learners (Teral 2007: 91, Torn 2004:

From a diachronic perspective, it is thus possible to recognise in the dialects of southern Italy a shift from an original equal sharing of features across both functional heads

Key Words: American Christians, Christian Right, Christian Zionism, US-Israel Relations, Conservative Christians Theology, State of Israel, Jews, Millennial beliefs,