Note from the editor: The following item is the first of
two articles on the Piedmontese Ballad, received from Riccardo
Grazioli. The second article will be published in the next number
of ITEM.
Diacritic signs have been realized in the following way:
Despite the constant use of constructions characterized by a maximum of verbal concision, the traditional Piedmontese ballad nevertheless possesses an outstanding ability to clearly evoke individuals, events and states of affairs while maintaining remarkable narrative incisiveness and intensity. In the Piedmontese ballad, in fact, we find a high degree of what might be called an "overproduction of sense", that is, where texts occasionally reveal very little, are often structurally fragmented and conceptually elliptical to the extreme, we find successful textual blending that is loaded with narrative impact. In other words, these are beautifully concise texts, that take the maximum possible advantage of a minimum of linguistic material.
In our view this felicitous equilibrium between verbal concision, referential transparency and narrative incisiveness can largely be considered the product of two specific factors: (a) the non-fortuitous use in the text of elements capable of guiding correctly, with a sufficient margin of unambiguity, the inferential effort that the audience must make in order to understand what is not stated explicitly in the text; (b) the generalized use in the texts of what the American linguistic tradition, associated with Labov, refers to as "evaluation" (Labov 1972, Wolfson 1979, Mirrer-Singer 1980, Mirrer 1987).
I have recently dealt at length (Grazioli ms.) with point (b) analysing aspects of the Piedmontese ballad such as the use of the gerund or the alternation between past and present verb forms in the sections of the texts without dialogue, while assigning to Labov's category of "evaluation" the frequently studied phenomenon regarding the insertion of elements extraneous to ordinary speech into the texts of the Piedmontese ballad.[1] Here, instead, I have concentrated on point (a), by analysing the use of vocative constructions in the Piedmontese ballad.
In section 1 I will introduce the theme of this paper by analysing some of the resources used to signal the presence of direct speech. In section 2 I will consider two uses of the vocative typical of the ballad: a) the vocative as a means of evoking textual objects whose existence is not made directly explicit by those sections of the text devoid of dialogue and b) the vocative as a means of orientating the hearer, and facilitating the understanding of the dynamics of dialogue --dynamics that would otherwise remain obscure, owing to the general lack of verbs of communication (and their relative logical subjects) in those sections of the text devoid of dialogue. In section 3 we will examine how these two particular uses of the vocative, used liberally in the Piedmontese ballad, can trigger off an entire set of inferences and thus serve as an authentic link in the narrative organization of the texts.[2]
From the mid-Seventies onwards the field of linguistics has devoted a great deal of attention to spoken language. Perhaps owing to the fact that the comparative aspects of the studies undertaken have often anchored the study of spoken language to that of written language, we now have at our disposal --albeit amid a sea of uncertainties and some occasionally very marked oscillations regarding the parameters of the object of study-- various analyses concerning the elements that typically mark the existing differences between these two distinct systems of communication: planning strategies, the circumstances of production, the use of particular codes and subcodes etc. (cf. Anderson 1990). Among these elements I will briefly discuss the aspect that arguably possesses the greatest degree of obviousness. I am alluding to the use, exclusive to writing, of punctuation, and more specifically to the use of punctuation as a means of informing the reader as to the boundaries of the direct speech contained within a narrative.
As we know, punctuation reveals a certain degree of functional overlapping with prosodic systems. But this overlapping, and the nature and the degree of importance assigned to the relationship linking the two systems, are variously viewed in the field of linguistics. The traditional tendency was to analyse punctuation as a highly imperfect and limited vehicle for the "rendition" and "signification" of prosody and the system of pauses.[3] More recently, and soundly argumented, we have begun instead to define punctuation as a system in itself, equipped with its own rules and analyzable _iuxta sua propria principia_, to the point that some researchers have tried not only to identify constants of use untrammelled by presumed prosodic input (see, for example, the various contributions contained in Cresti et al. 1992), but also to formulate fragments of authentic "punctuational grammars" (Nunberg 1990). From whatever point of view one considers the interferences linking the system of punctuation and that of prosody, and whatever the degree of postulated affinity or concomitance might be, it nevertheless seems indubitable that punctuation and prosody, precisely because they are systems, function in accordance with non-comparable principles. Punctuation, at least as far as it is commonly used, is configured as a yes/no-type system, based on the presence or absence of a given element. Prosody, on the other hand, seems to allow for a theoretically infinite range of possible realizations in relation to its components. This leads us to an important characteristic of punctuation: the absolute unambiguity that some of its symbols, in contrast with the prosodic elements that, rightly or wrongly, have been traditionally considered their counterparts, can assume within the communicative economy of a text.
Consider for example the use of European quotation marks, or various other punctuation marks that, according to the conventions of writing or typography used, indicate the presence of direct speech in a written text. In particular contexts they can perform their delimitative function with absolute transparency. In this way a written narrative, expressed through the representation of direct speech between two characters, not only produces constructions of the type shown in (1) below, where the revelation of the character's identity is entrusted to the subject of the verb following the speech enclosed within the quotation marks,
(1)
where the simple alternation of the utterances --given a context capable of evoking the existence of two individuals and of providing the identity of the speaker that begins the sequence-- makes it possible in any specific fragment of text to eliminate any doubt as to the identity of the speaker and the hearer. With regard to an interpretative model intended to link the utterances present in the communicative exchange to the individuals evoked by the context, a series (2) sequence supplies only one possible inferential path. If speaker A's utterance is the first to be enclosed between quotation marks, then speaker B's words must be enclosed within the second set (of quotation marks), then speaker A will provide the third, and so on.
Now let's shift our attention to oral narratives: how does the audience know where there has been a change of speaker? Obviously, there are numerous ways of doing this and some of them often appear together. If we limit ourselves to strictly oral devices we can, for example, quote the use of particular prosodic or paralinguistic traits that characterize the utterances of a given speaker (Lavagnino ms.), or the use of various combinations of pause and rhythm (Keyworth 1979; Kovacs 1981) or again --analogous to what in monologues appears to serve to divide up the text into tonal "paragraphs" and "subparagraphs" (Lehiste 1982; Hirschberg and Pierrehumbert 1986)-- the increment of the fundamental frequency localized in correspondence with a point in the text where we find an utterance made by a person who is not already speaking (Lavagnino ms.). In the traditional oral narratives of the area from which our corpus comes, however, the primary mode of signaling the presence of direct speech is reduced to the insertion of a verb of communication associated with a specific speaker and in some cases with the individual to which the utterance is directed. The default verb of communication is the neutral one "to say," which is probably preferred to verbs such as "whisper," "shout," "order," etc., in relation to the habitual intonational qualification supplied by the narrator to the speech act attributed to the logical subject of the verb. The specifications may be effected by using proper nouns, common nouns and pronouns, or they can be nullified by recourse to the use of ellipsis.
For instance:
[Arietti 1888: 424]
Na volta a jera un re ch'al avia un fieul ch'al era un crin, e sto crin dunque a spasegiava par j apartament e d'tans an tan a l'era cativ e a fasia d'espr[e/]si. So pare un di caresandlo ai dis: Cos'l'astu chit ses cos[i\] cativ, cos' l'astu? Chiel a s'e butasse a fe ii vers ca fan i crin: Eu, eu, i veui mari[e>]me, i veui la fija del panat[a\]. [Once upon a time there was a king, who had a son, and this son was a pig, and this pig was kept in the house, as if he were a boy, and sometimes he behaved very badly. Hugging him one day his father said: "What's the matter with you? Why are you so naughty?" And he answered the way pigs do: "Oink, oink, I want a wife, I want a wife, I want the baker's daughter."]
As far as this fundamental method of rendering direct speech is concerned, the problem regarding the identification of the alternation of the speakers and hearers along the axis of the narrative is therefore reduced to the problem of resolving anaphors and ellipses, and as such it could be tackled, without any need to fall back heavily onto inference, by an interpretative model analogous to that elaborated by Sidner (1983), Carter (1987) and Carberry (1989). Note, moreover, how this method of rendering direct speech is not, as far as the Piedmont area is concerned, the exclusive property of metrically non- formalized speech. Virtually all the narrative texts produced by the street singers or _cantastorie_[4] included in V1 and V2 systematically use this method:
["Il lamento del soldato Parente Giuseppe che si sfida a duello col tenente", V2:93-94]
[Beppino] Si parte dal quartier allegramente deciso proprio di trovar l'amante e percorrendo incontro al suo tenente che disse: andate indietro all'istante Disse Beppino: indietro [non] torner[o\] rispose il tenente: vi fo' morir in prigion Fate silenzio non fate il monello presto ti mander[o\] alla reclusione Beppino si volt[o\] veloce e snello e disse: almeno mi dica la ragion! Tosto il tenente la spada impugn[o\] lo stesso fa Beppino e la lotta principi[o\] Parevan due leoni alla foresta dicendo l'un all'altro: traditore! ferito fu il tenente nella testa dal giovinotto sull'et[a\] del fiore [Gaily he [Beppino] left the barracks resolved to meet his lover; but on his way he met his lieutenant who said: "Turn back immediately." Beppino said: "I will not turn back!" The lieutenant answered: "I'll see you rot in jail! Be silent and don't play the fool, or soon behind bars you will be!" Beppino turned swiftly to face him and said: "At least tell me why!" Boldly the lieutenant unsheathed his sword, as did Beppino, and the duel began. They looked like two lions in the jungle, while one to the other "Traitor!" said. The lieutenant was wounded upon the head by the young man in the prime of his youth.]
Now we come to the ballads of our corpus. First we should note that they are completely permeated by direct speech: in the entire corpus indirect speech appears only very rarely, and texts that have no communicative exchanges at all are just as rarely found. In addition we ought to note that, despite the pervasive use of direct speech, constructions implying a verb of communication in our ballads, constitute the exception, and not, as they are in the spoken or sung narratives of the _cantastorie_, the rule. On the other hand, the few constructions based on verbs of communication to be found in the traditional Piedmontese ballad are accompanied by the fact that their use is substantially dependent upon distributional terms as well as details that appear to indicate functions that are not restricted to the mere indication of direct speech. Indeed,
(1) verbs of communication are often used in set contexts, for example:
a. When the communicative exchange occurs between a group of individuals that at a given moment in the story functions as a single narrative character.
[TAV:56] poi sa di[s.][i\]u fra lur tre: duv'la andarumni a suter[e\]?[5] [And then they said, all among them three: "Where shall we bury her?"]
b. When the direct speech is uttered by collective bodies which are extraneous to the main plot and whose role in the text is restricted to an evaluation of the characters or their actions.
[N:43H, 17-18] Dame bele a le finestre e signur a li balcun, A cr[i/]an: --Viva la 'rgina, la spuza d[e:]l nost padrun!-- [The fair ladies at the windows, the lords at the balconies, cheer: "Long live the queen, our lord's bride!"]
c. When a character prefigures a series of events or a state of affairs set within a frame that is heavily influenced by the classic topoi of the ballad genre; in this case the verb takes the future tense.
[N:27A, 22-24] Pi[e\] me corin bel, port[e\]-i-lo a Margherita; Chila a lo pier[a\], chila a lo bazer[a\], Dir[a\]: --Che gran mal[o:]r, l'[a\]n f[a/]it m[u:]r[i\] me c[o:]r!-- [So take my little heart, carry it to Margherita; she will take it, she will kiss it, she will say: "What a misfortune! They have let my love die."]
d. When cases b and c are found in concomitance with each other.
[F2:85,19-20] Titt ra gente chi passran-nhu, Diran: cs'[e\]le ista matin? [And all the people who will pass by, will say: "What has come about this morn?"]
e. When speech is put into the mouths of individuals who, in the real world, do not speak. In these cases it appears that the verb of communication serves explicitly to assure the audience of the plausibility of a state of affairs that would otherwise be rejected by common sense. Typical cases would include "children only a few months old"
["dona bi[e:]nca", V2:9] Un fantolin di sei mesi al a rispund[u:] pap[a\] b[e:]iv p[a\] lul[i\] che la mia i mamma vol fate morir [An infant of six months replied: "Daddy, don't drink that, for my mother would kill you."]
and animals.
[F2:97, 3-4] Ra firmija sa i dis: --V[o>]ti spus[e\]me mi?-- [The ant says: "Will you marry me?"]
f. When the introduction to the direct speech is delegated to a formulaic expression based on contraposition.
[N:17A, 7-9] Gentil galant va a la ceza, a l'[a\] dumand[a\]-la a [a/]uta vus, A [a/]uta vus a l'[a\] dumand[a\]-la; a bassa vus a j'[a\] rispus: --Cul anlin ch'i l'avei spuz[a\]-me, guard[e\]-lo s[i\] ch'l'[a/]i ant [e:]l dil. [The gentle gallant went to the church, loudly he asked for her loudly he asked, softly she answered: "Look here, our little wedding ring is on my finger."]
2. Constructions including verbs of communication are normally used - -if used-- not singularly, but in clusters. In fact, it is reasonable to suppose that within the Piedmontese ballad there is a compositional tradition, albeit a decidedly minor one, which often (but not exclusively) falls back systematically on this means in order to indicate the insertion of direct speech in a narrative. These constructions may occur in contiguous verses
[F2:51,3-6] Lo soi mar[i\], su i ha ben dicc: --Ma cosa v[o>]ti mai marcant[e\]e?-- Aluizina a i rispund: --Marcantirumma reif e savun.-- [Her husband spoke clear: "But what shall be our trade?" Aluizina answered him: "Our trade shall be selling thread and soap."]
or at a distance, and appear with a notable frequency, although never with the frequency typical of spoken narratives or of broadside ballads.
[F1:XXII, 5-28] Soi pari i j dis: --La bela, T' veuli mni a Genua con m[i\]?-- [...] So pari i j dis: --O bela Cosa t' tacca da cant[a\]?-- [...] La bela dis al so car pari: --O chi mai j han da mass[a\]?-- E so pari a j dis: --La bela I han da fati muri t[i\], [...] E i [i\] cumensa a d[i\]: --La bela, Mi volivi spos[a\] mi?-- [Her father said: "Fair damsel, will you come to Genoa with me?" [...] Her father said: "Fair damsel why are you singing?" [...] The fair damsel said to her beloved father: "Who shall be executed?" And her father said: "Fair damsel, 'Tis you that shall executed be" [...] And so he [the executioner] begins to say: "Fair damsel, will you marry me?"]
3. Constructions including a verb of communication seldom occur in non-marked forms such as "A says to B." More often these constructions contain:
a. Rather than "to say," a verb that is semantically less neutral and better suited to specify or emphasize the action.
[F1:XXIX, 13-15] E cul spessiari s'butta a cri[a\]: --Sun gnanca ancura marid[a\], E gi[a\] la balia bsogna pag[a\].-- [The spice merchant begins to cry: "I'm not even married yet, and already I must pay a wet nurse."]
b. An adverb such as "bin" or "ben," used to highlight the verb phrase (cf. Terracini 1957a:206); in Nineteenth-century collections "dis bin", "ben dicc", "ben d[i\]-je", etc., are indeed the nucleus of standard constructions based on verbs of communication.
[N:143A, 9] Signur lo re s'a j'[a\] bin dit: [...] [The lord king said clearly: [...]]
c. A verbal periphrasis that makes plain the particular aspectual significance (inchoative, durative, etc.) of the action.
[N:51B, 17] S'a s'[e\] but[a\] crie: --Oim[e\]! ch'i sun tradia. [She began to cry: "Woe is me for I am betrayed."]
Now let's return to examples (1) and (2), which illustrated the two ways of rendering direct speech on the printed page. When written down, 90% of the direct speech contained in our ballads would fit within the parameters of scheme (2). And yet, being oral texts, ballads cannot provide the punctuation marks that delimit direct speech and eliminate both ambiguities relative to the extent of the spoken phrases and to the identity of the speakers. Moreover, because they are sung, they cannot supply the audience with prosodic signals analogous to those that may be used in spoken narratives. This, therefore, leads us to the question: what mechanisms do our ballads employ to permit the audience access to the relevant information required in order to understand the locutional relationships in the texts? The answer to this question necessarily involves a reference to "cooperative activity" --an entirely pragmatic notion-- that leads the receiver to infer from the text that which the text, on a superficial level, does not express overtly, but supposes and implies. As they only make minimal use of verbs of communication within the narrator's speech, our ballads must employ elements from within the direct speech of the characters. The audience, therefore, must make the interpretative effort that, in a sufficiently unambiguous fashion, manages to make explicit those locutional relationships that are not expressly sanctioned by the text lacking in utterances of the type "A says to B."
On the subject of communicative exchanges produced by two interlocutors, let us consider the following fragment:
[N:14A, 7-8] --Coza dir[a\] la mama mia, che n'a sto tant a riturn[e\]? --Pens[e\] pa pi a la vostra mama [...] ["What shall my mother say, since I tarry so?" "Of your mother think nomore [...]]
Here, in order to assign the status of dialogue and the identity of the interlocutors, as well as establish who is speaking and who listening, an interpretative model might roughly go as follows:
["donna" says to "pretendente":] --Coza dir[a\] la mama mia, che n'a sto tant a riturn[e\]? ["pretendente" says to "donna":] --Pens[e\] pa pi a la vostra mama [...]
In general, however, the interpretative paths that the Piedmontese ballads oblige the audience to follow in order to re-establish --on the basis of that which is contained in the direct speech-- those things that the narrator does not make explicit, are far less burdensome than the one summed up in points (i-iv). The details of the locutive situation within the text, in fact, can usually be obtained from paths based on more direct and explicit elements, i.e. on vocative constructions. In focalizing an hearer, a vocative does not only indicate that we are in the presence of direct speech, but also may serve to arrive at a substantial formative basis that permits a complete identification of the organization of the dialogues within the text. It is precisely this enormous informative power of the vocative that allows us to explain its pervasive presence in the Piedmontese ballads, a presence well exemplified by this fragment:
["Ceciliota", V1:120] Eran tr[e\] madamig[e/]le ch'a veniju da Liun Cecilia la pi b[e/]la 'l [a\] 'l so mar[i\] 'n par[s.]un O, bund[i\], sur capitanne 'l [e/] 'na grasia che mi vur[i\] 'l [e/] 'na grasia che vurija 'l [e/] la vita al me mar[i\] O, s[i\], s[i\], bella Cecilia o s[i\] s[i\] b[e\]n p[e:]r lul[i\] baster[a\] una sola noite a venir durmir cun mi Andar[u\] a la par[s.]un andar[u\] a dilu al me mar[i\] e se chi[e\]l sar[a\] cunt[e\]nt e cunt[e\]nta sar[u\] mi So mar[i\] dalla finestra da tant luntan 'l [a\] 'vdula 'vn[i\] o, che n[o:]ve 't porte Cilia o, che n[o:]ve 't porte 'd mi [There were three damsels who came from Lyon. The most beautiful was Cecilia, whose husband was in prison. "Good-day to you, Sir Captain. May I ask you a favor? The favor I ask of you is the life of my husband." "O yes, yes, lovely Cecilia! O yes, yes and for that! It would be enough that you spend only one night with me." "I shall go to the prison and tell my husband and if it pleases him I too shall be pleased." Her husband from the window saw her coming from a great distance. "O what news do you bring, Cecilia? O what news do you bring to me?"]
The vocative, however, is systematically used in our corpus also as a way of establishing new textual objects, a fact that is shown to be of extreme importance regarding the linguistic characterization of the texts since it allows us to make a clear distinction between the constructive strategies belonging to the traditional spoken narrative (e.g. folktales) and those of the ballad.
In folktales, the introduction of the protagonists of the action is usually reserved to there-insertion sentences situated at the beginning of the text. These are an integral part of what Rumelhart (1975) describes as "setting": one or more statements that orient the receiver regarding the characters and the space-time coordinates of the action, and by so doing initialize the discourse model that lists the objects, the properties, relationships etc., that have a part to play in the discourse. E.g.: "Once upon a time, in a far away land, there lived a good king, his beautiful queen and their daughter, Princess Cordelia [...]".
As a matter of fact, in the Piedmontese ballad, such
functional statements are quite frequently used to place the
characters within the discourse frame.
Indeed, in this way some ballads begin as folktales do:
[N:53A, 1-3] Ant la vila de T[u:]rin, l[a\] s'a j'[e\] tre bele fije, Bianche e russe cum' na fiur. J'[e\] tre giuvo capitani, ch'a n'a van f[e\]-je l'amur. [In Turin town there are fair maidens three. Fair [literally: "white and red"] as flowers. There are three young captains who would court them."]
But very often the ballads of our corpus do not possess any introduction provided by the narrator, and so they form a macroscopic example of an extremely strong tendency to adopt language that is openly and unconditionally dramatic. In such cases the ballad narrative begins ex abrupto with a section in direct speech including a vocative. Indeed, in some collections included in our corpus, this sort of opening emerges as the most commonly used of all, e. g.,
[S2:"Girometta", 1] "Girometta de la montagna, turna al to pais ["Girometta from the highlands, go back to your village];
[S1:4, 1-2] <<O d[i\]me'n po', bel giuvo da capel burd[a\] / Sav[e/]ise d[e\]me d'n[o:]ve d[e:]l me inamur[a\]?>> ["Tell me, young man with the brimmed hat / can you bring me news of my beloved?"];
[CEC 10-1A 1.2, 1] <<V[o:]ste ven[i\]r, ti, o d[o/]nna bianca? V[o:]ste venire a spaso con mi?>> ["Will you come, white lady? Will you come walk with me?"];
[CEC 11-2B 15.1, 1] <<C[o/][s.]a f[e/]sti, o l[i\], berg[e/]ra, larg[a\]nd i vost mut[u\]n?>> ["What do you there, shepherdess, while tending your sheep?"];
[CEC 15-6A 40.2, 1] < <O di[s.]imi 'n p[o/], vui mama, Fiurentina duv'[e/]lu 'nd[a\]?>> ["Tell me, o mother, where has Fiurentina gone?"].
In these cases, the vocatives, present in the opening utterances of the text, could very well be said to carry out the default communicative function of the vocative, i.e. that of attracting the interlocutor's attention. In fact, from the point of view of the various speakers using the allocutive forms, these vocatives serve as attention-getting devices (cf. Haverkate 1984: 68 ff.). From a textual point of view, however, these vocatives carry out functions elsewhere that belong to "setting." They allow the initialization of a discourse model that comprises at least two textual objects (the speaker and the addressee), and by virtue of the specification of these two objects supplies the basis for the development of the story. Consider, for example, the following fragment:
[N: 14A, 1-3] --O marinar de la marina, o cant[e\]-me d'[u:]na cansun. --Munt[e\], bela, s[u:] la mia barca, la cansun mi la canter[o\]. --Quand la bela l'[e\] st[a/]ita an barca, bel marinar s'b[u:]ta a cant[e\]. ["O sailor from the sea, sing a song to me." "Step aboard my ship, fair lass, and I will sing the song." When the fair lass was aboard, the handsome sailor began to sing.]
The vocative "O marinar de la marina," part of v. 1,
implies that the utterance here contained is addressed to
someone. The existence of this vocative therefore indirectly
allows the initialization of a discourse model that, in relation
to the first utterance in the fragment, comprises among other
things:
a) a representation of a sailor X;
b) a representation of an individual Y about whom we know only
that he/she has made a request addressed to the sailor X.
On the other hand, v. 2 contains the vocative "bela,"
an expression used generally and generically in the Piedmontese
ballads to identify female characters. So v. 2 can be taken as a
reaction to the discursive stimulus (in particular a
"request") supplied by Y in v. 1. And given that there
is nothing that leads us to suppose that in the locutive
situation within the text there exist more than two speakers, it
is possible at this point to update the discourse model of the
fragment by associating the individual Y --already present in the
model, but as an individual devoid of an identity-- to the
identifier "bela." By the time we arrive at v. 3 our
discourse model is already complete with representations of the
two characters of the ballad, and it is no accident that it is
precisely in v. 3 that they are mentioned with definite noun
phrases ("la bela" [the fair lass] and not "[u:]na
bela" [a fair lass]). In v. 3 therefore, we find ourselves
in possession of the information that could have been introduced
into the text elsewhere in a declarative form (e.g. "There
is a fair lass. There is a sailor. The fair lass sees the sailor.
The fair lass says to the sailor: <<...>>. The sailor
says to the fair lass: <<...>> [...]). The
"setting" is in fact supplied implicitly by the
dialogue and the vocatives, which bring the discourse model of
the fragment to a stage that is already more advanced than the
point at which Rumelhart (1975) fixes the insertion of a
"stative proposition" as a sign of the closure of the
introductory section of a story (e.g. "One day, as Princess
Cordelia was walking near the palace [...]"). And that is
not all. Not only are the propositions "there is a fair
lass," "there is a sailor," etc., expressed
without being explicitly asserted (as they can be inferred
starting from vv. 1-2), but dramatic action is developed right
from the start, thus inbuing the text with greater incisiveness
and narrative impact.
As mentioned above, the introduction of the characters effected by the use of a vocative is a common practice. This in itself would seem to exclude the idea that the device is not an organic part of the set of linguistic and rhetorical tools used by performers to create and recreate the texts. It is nevertheless interesting to note, in order to emphasize the importance of this technique, that --as far as the Piedmontese ballad is concerned-- it is by no means rare to find that the text contains characters, devoid of a precise identity, whose sole or principal purpose is to pronounce vocatives capable of establishing the principal characters. For example, this is the case in N:54A, where the initial direct speech is uttered by an individual who does not appear in the text thereafter and who has therefore exhausted his usefulness by the time he has established the "fieta," attributed to her the property of not being married and invited her to sing
([N:54A, 1] --O cant[e\], cant[e\], fieta, finch[e\] sei da marid[a\]. ["Sing, sing young girl, while you are not yet married"])
, or it is the case in F2:27, where the initial line delivered by the individual who does not appear again in the text serves to solicit a response on the part of the "gentil galante" established by the vocative
([F2: 27, 1-2] --Duv'and[e\]v, gentil galante, / Duve s[i\]ve ancamin[a\]?-- ["Where are you going, gentle gallant, / whither would you go?"]).
It is also the case in N:99A, where the initial line serves not only to establish the individual "Roza," but also to offer her an invitation capable of orienting the audience with regard to the topic of the text
([N:99A, 1] --O Roza, b[u:]ta ij fiure, ch'a t'[a/]i d'and[a\] bal[a\]. -- ["O Roza, deck your dress with flowers, and you shall go to the dance."]).
Dialogues in the Piedmontese ballad are always delivered face
to face.
This means that the interlocutors verbally interact at the same
time and in the same place. And this, in a type of narrative text
where time and space are determinant factors for the construction
of the text, has some very interesting side effects.
Consider, for example, the following fragment:
[N:23D, 1-4] --Giuvinotto, bel giuvinotto, duva s[e/]i-ve ancamin[e\]? --Vad a caza de la vidovella, ch'a gh'[a\] na fia da marid[e\]. Vidovella, bella vidovella, la vostra fia m'la vorr[i\] d[e\]? --La mia fia l'[e\] ancor piciota, non [e\] ancor buna da marid[e\]; ["Young man, handsome young man, whither are you bound?" "I am going to the widow's house, she has a daughter I would fain wed. O widow, gentle widow, will you give me your daughter?" "My daughter is too young, too young to be wed."]
In v. 1 of this fragment, an individual (who does not appear again in the narrative) introduces with a vocative the individual referred to as the "bel giuvinotto." In v. 2, the "bel giuvinotto" in turn introduces an individual referred to as the "vidovella." Finally, in v. 3, we find the vocative "Vidovella, bella vidovella," on the basis of which the audience can consider v. 3 as direct speech which, as it has focalized the individual known as the "vidovella," may plausibly suggest that the speaker is the "bel giuvinotto." In this way, given that in order to hold a dialogue in the Piedmontese ballad two interlocutors must find themselves face to face, the audience can infer: a) that by v. 3 the "bel giuvinotto" has already covered the distance that in vv. 1-2 separated him from his goal, i.e. the house of the "vidovella"; b) that the narrative time in which the communicative exchange contained in vv. 3-4 occurs is posterior to that of the communicative exchange contained in vv. 1-2.
Note that all this occurs without an intervening narrator making explicit assertions of the type "the 'bel giuvinotto' has walked as far as the widow's house; he has knocked at the door; the widow has opened, etc.," nor has there been any necessity to fall back on even more detailed description. Everything is delegated to direct speech and, in the final analysis, to the vocative "Vidovella, bella vidovella," without which the locutive situations of the text and the corresponding locative and temporal specifications would not be clear.
Now let us consider the following example, similar to the preceding one, but with a much more complex organizational structure:
[N:96A.I, 1-18] J'[e\] [u:]n giuvo an custa vila ch'a v[o:]l prende muj[e\]; Ma sul che na noitea n'a va d[u:]rm[i\] cun l[e\]. S'a ven meza noitea, 'l galant a s'[e\] lev[e\]; A pia lo so capelo che via na v[o:]l and[e\]. --Mar[i\], caro marito, duva vui v[o:]li and[a\]? --Mi vad a l'osteria, e vui mi v' lasso a c[a\]. Prunt[e\], madama l'osta, da b[e/]ivi e da mang[e\], Mi andar[o:\] pi[e\] la dota, mi v' porter[o:\] i den[e\]. -- La p[o/]vera spuz[e:]ta 'l matin a s'[e\] lev[e\]. N'[a\] pi[a\] la sua ruch[e:]ta, a c[a\] sua na v[o:]l and[e\]. --O bund[i\], pare e mare; mi sun turn[a\]-m-ne s[i\]; Sun vn[u:]a a d[e\] dle n[o:]ve d[e:]l me caro mar[i\]. --T'l'avio sempre d[i\]-t-lo, ch'a l'era [u:]n gi[o:]gadur; E ti t'as rispond[u:\]-ne, ch' l'era to prim amur. Va p[u:]ra, la mia fia, va p[u:]ra a la tua c[a\]. Che t'[a/]bie [u:]n po' passiensa se ti bastuner[a\]. --Mar[i\], caro marito, avei-me cumpassiun, Bat[i\]-me cun le mani, e lass[e\] gi[u:\] 'l bastun.-- [In this town there lived a young man, who took him a wife; but laid with her only one night. At midnight the young gallant rose from his bed; and put on his hat, for he would go away. "Husband, dear husband, where would you go?" "I am going to the tavern, and you will stay home." "Bar maid, bring me wine and meat; and with my dowry gladly I will pay you!" The sad wife rose the next morning. Taking her distaff she returned to her parent's home. "Good day father, good day mother; I have returned, with news to tell of my husband dear." "I told you and I warned you that he was a gambler; yet you replied that he was your true love. So go back, my daughter, go back to your home, you shall learn to bear pain, when he beats you." "O husband, dear husband, have some pity on me: please beat me with your bare hands, and leave the stick alone."]
Here the understanding relative to the various temporal breaks that occur as the story develops, and the various movements through space carried out by the characters is wholly delegated to the audience's capacity to infer that the aforementioned movements through time and space all spring from the presence of vocatives in the text. The vocative "Mar[i\], caro marito" simply indicates that v. 5 is direct speech delivered by the wife to the husband. By the time we get to v. 5, therefore, husband and wife are in the same place. The vocative "madama l'osta" indicates that in v. 7 the husband has gone to the tavern and therefore that between v. 6 and v. 7 a temporal break has occurred. The vocative "pare e mare" indicates that the wife, from the place in which she was in v. 5, has moved to the place where her parents are by the time we get to v. 11; a temporal break is therefore implicit in this case too. On the other hand, the vocative "la mia fia," contained in v. 15, indicates nothing as far as times and places are concerned; more simply it can be considered a means of clearly marking the fact that this is a continuation of the direct speech begun in v. 13. The vocative "Mar[i\], caro marito" in v. 17 indicates a further temporal break and a further change of place. But here the localization of the action lends itself to competing interpretations: in v. 17 either we are in the tavern (and in this case only the woman has moved) or we are, more plausibly, in the place where the first dialogue occurred (in this case both the woman and the husband have moved).
Now, in the Piedmontese ballads, this way of implicitly evoking the spatial dislocation of the characters and the presence in the text of temporal breaks is very widely diffused. However, it should be emphasized that the side effects of the use of vocatives are not exclusively limited to questions of narrative time and space. Causes and motivations also come into play. In determined contexts, the use of a vocative can in fact trigger inferences that --possibly based on common frames-- manage to define possible courses of events responsible for a state of affairs that is revealed at a given point in the story. With this in mind let us consider the following fragment:
[N:103A, 7-11] (Context: a young girl makes love with a boy without her parents' knowledge. On returning home she tells her mother that she does not feel well. Her mother understands what has happened and obliges the girl to go to the judge:) --Se t'sei t[u:]ta malada, mi sai che mal ch'a l'[e\]. Duman matin bunura dal gi[u:/]des t'andar[e\]. Ch'a senta s[i\], sur gi[u:/]des, ch'a senta mia razun. A m'[a\]n baz[a\] la biunda; na v[o:]i sodisfassiun. --O sent[i\] s[i\], cul giuvo, vui s[e\]-ve cundan[a\]. ["If you feel ill, I know what ails you. Early in the morn to the judge you will go. O hear me, lord judge, hearken to my plea. They have kissed my blonde lass and I would have satisfaction." "Hear me, boy, for you are condemned."]
In this fragment the vocative "sur gi[u:/]des,"
inserted in v. 9, shifts the action from the girl's house to the
place where justice is carried out.
But there is no mention of _whom_ it is that has moved. Certainly
it is one of the girl's parents (the individual who pronounces
the vocative "sur gi[u:/]des") and perhaps, given the
command contained in v. 8, the girl. But from the presence of the
vocative "cul giuvo" in v. 11 one understands that the
boy too has been ordered to present himself before the judge.
This can evoke a sequence of actions --not explicitly referred to
in the text, and in any case indispensable in order to justify
the presence of the young man in the place where justice is
carried out-- that implies that the girl's parents have sought
out the young man and obliged him to present himself before the
judge, etc.
In this contribution we have shown how the vocative is used in
the Piedmontese ballad to convey information of great importance
and with extreme precision and incisiveness. In particular we
have emphasized how in determined circumstances a vocative may,
by functioning as an informative shortcut,
On the basis of this evidence it can be argued that the vocative can carry out an important role with a view to the linguistic characterization of the general narrative strategies adopted by the Piedmontese ballad. In addition, the attention that other scholars --albeit for reasons and with intentions that differ from ours-- have dedicated to the use of the vocative in the Spanish "romancero" (Martin 1989), in the English and Scottish ballads (Kekäläinen 1981: 55 ff.) or in Romanian narrative songs (Renzi 1969: 30 ff.) lead one to believe that the vocative may turn out to be a useful element in the linguistic characterization of the narrative strategies adopted by a good number of European ballad traditions.
#. Paper presented at conference "Ballads and Boundaries: Narrative Singing in an Intercultural Context", June 21-24, 1993, UCLA, Los Angeles
1. For example the particle _s_ (Terracini 1957b, Spitzer 1958) in contexts of the type "S'a l'[a\]n pi[a\]-ro, l'[a\]n li[a\]-ro, l'[a\]n mn[a\]- lo 'nt la tur d'Paris" ("they captured him, they bound him, they took him to the tower of Paris"), or the particle _d_ (Salvioni 1902, Terracini 1957b: 91 ff.) in contexts of the type "j'[e\] d'ina barb[e\]ra" ("there is a woman barber").
2. Here follow the sources used in the article:
CEC: a collection including approximately 300 ballads
recorded on tape by various researchers during the years
1960/1988. The collection, edited by the "Centro Etnologico
Canavesano," is for the most part made up of hitherto
unpublished material.
F1: G. Ferraro, _Canti popolari monferrini raccolti ed
annotati dal Dr Giuseppe Ferraro della R. Scuola Normale Sup. di
Pisa_, Torino- Firenze: Loescher, 1870.
F2: G. Ferraro, _Canti popolari del Basso Monferrato
raccolti ed annotati da Giuseppe Ferraro_, Palermo: Luigi Pedone
Lauriel, 1888. N: C. Nigra, _Canti popolari del Piemonte_,
Torino: Loescher, 1888.
S1: R. Sinigaglia, _36 Vecchie canzoni popolari del
Piemonte_, Milano: Ricordi, 1957.
S2: R. Sinigaglia, _24 Vecchie canzoni popolari del
Piemonte_, Milano: Ricordi, 1958.
TAV: The "Teresa Amerio Viarengo Collection."
The collection, comprising about a hundred titles, some of w hich
are available in several variants, was recorded on tape by F.
Coggiola and R. Leydi between 1963 and 1965 (cf. Leydi 1993). To
this day only a minimal part of the material has been published.
V1: Vigliermo, A., _Indagine sul Canavese. Trascrizione
dei testi e nota fonetica di A. Genre_, Ivrea: Priuli &
Verlucca, 1974.
V2: Vigliermo, A., _Canavese che canta_, Ivrea: Priuli
& Verlucca, 1986.
In the course of this paper, in order to indicate the texts contained in the published collections, we shall refer, whenever possible, to the progressive numeration supplied by the publication (e.g., text no. 1, in version A, contained in collection N will be designated as N:1A, and followed by an indication of any verses quoted), or, if a reference to a numbered order is not possible, we shall refer directly to the title assigned by the editor of the collection and to the page number (the text "La Franc[e/]sa", published on page 26 of V2 will therefore be referred to as "La Franc[e/]sa", V2:26). To indicate the texts contained in CEC we shall use the code numbers contained in the booklet _Costantino Nigra, Canti popolari del Piemonte. Cent'anni dopo_, printed by the Centro Etnologico Canavesano in 1988, which supplies the complete index of the collection as well as information regarding the informants, field collectors, places and dates of the surveys. To indicate the texts included in the TAV we shall make use of the code numbers used in the index prepared by E. Bergomi in view of the forthcoming publication of the collection.
Regarding the reproduction of the published texts, we have followed, with one exception, criteria of absolute fidelity to the original transcriptions. In other words, we did not feel it suitable to standardize the spelling (which differs considerably depending on the editor) nor did we intervene when we found obvious inaccuracies or incongruencies in the transcription (typically these were problems related to the rendering of pronominal accumulations, to which editors often failed to attach due importance). The exception concerns the verbatim repetition of entire verses, which were eliminated inasmuch as they were considered irrelevant for the purposes of this study. The transcriptions of the unpublished texts included in the CEC and TAV are by E. Bergomi.
3. One emblematic example of this is the stance taken by Markwardt (1942): "Punctuation is to a large degree a system of conventions whose function is to assist the written language in indicating those elements of speech which cannot be conveniently set down on paper: chiefly, pause, pitch and stress."
4. For the broadside texts, usually composed in Italian and not in dialect, diffused by the _cantastorie_ in the public square and not linked to any specific strictly regional tradition, see among others Pianta 1982: 99 ff. and Hirdt 1978.
5. Incidentally, it is interesting to note how expressions of
the type "a na dizio tra lur tre," used for
communicative exchanges between the members of a group, are
revealed as crystallized verbal routines and, apparently,
partially emptied of their semantic content. What happens in fact
is that such expressions are used ritually, in a textually
incongruous fashion, even when the context does not justify their
use.
This is the case in the following verse, inserted in a text that
does not supply plausible antecedents for "lur tre":
[N:6C, 15] A na dizio tra lur tre: --Che bel prezent l'um-je mai da f[e\]? [And so they said, all among them three: "What present shall we offer him?"]
or the first verse of this strophe,
[TAV:36] P[o:]i sa di[s.][i\]u fra lur autri: <> [And so they said, all among them: "I have a brother in France, and did he know I were being held here, he would burn Susa town, and condemn to death even the judge."]
which reveals evident incongruencies with regard to the direct speech that follows.
Riccardo Grazioli
Archivi della Comunicazione Orale
V. Melchiorre Gioia, 121
20125 Milan
ITALY
Archivio Storico:- ex Dipartimento di Musica e Spettacolo - Universita' di Bologna