SOURCE: Andrews, Richard. “Improvised Comedy.” In Scripts and Scenarios: The Performances of Comedy in Renaissance Italy, pp. 169-99. Cambridge: Cambridge University Press, 1993.

[In the following excerpt, Andrews discusses the bases for improvisation in the commedia dell'arte, providing examples of the short and long frameworks in which actors could create scenes.]

DEFINITIONS AND EVIDENCE

In 1567, the Duchy of Mantua was visited by two competing theatre companies, both including women: one was actually directed by an actress whose stage name was ‘Flaminia’, and the other run jointly by a ‘Pantalone’ (possibly Giulio Pasquati) and the actress Vincenza Armani. The artistic and commercial rivalry between the groups was made more interesting for the public by the fact that each leading lady was being courted by a different aristocratic patron—it is reported that the whole city was divided between fans of ‘Flaminia’ and supporters of Vincenza. As well as mounting improvised comic scenarios, each woman also starred in a more serious play, one based on the Virgilian story of Dido and the other taken from Ariosto's Orlando Furioso.¹ By the 1560s, then, professional companies were an established fact of life in northern and central Italy. The first surviving notarial document regarding the constitution of such a group dates from 1545.²

It is not intended here to give a comprehensive presentation of commedia dell'arte—that is a task which has already been performed by various reliable scholars,³ and readers who wish to acquire a full picture of the elements of the genre should turn to them. In this chapter we shall be pursuing only those aspects on which there may be something new to say, and which arise out of what has been presented so far: namely the links, in so far as they can be determined, between dramaturgical practice in scripted and in unscripted comedies. However, the very mention of ‘dramaturgy’ in relation to theatre which is ‘unscripted’ makes it clear that there will have to be some justification for what we are about to propose. This in turn creates a need for a selective factual outline. In describing a phenomenon which lasted over two hundred years, preference must be given in the present context to what seems to be true for the earliest period. Scholars now tend to identify a ‘classic’ (some even use the term ‘heroic’) phase which ends around 1625.⁴

The term commedia dell'arte as such is not found in print until it was used by the Venetian comic dramatist Carlo Goldoni in 1750⁵—the phrase may already have been current in speech. The word arte in this context refers simply to artisan skill, without the inflated pretensions now associated with ‘art’; and in the sixteenth century itself it was also the term for a trade or professional organization, a guild. In theatre history we now use the expression commedia dell'arte to refer to a very particular phenomenon: a theatre based on the use of ‘masks’ (a term which we must then define rather carefully), and in which plays were improvised on the basis of an outline scenario, rather than being learned word for word from a script. With these two essential characteristics—‘masks’ and improvisation—commedia dell'arte is eventually found in countries other than Italy, though its Italian origin is always recognized: within Italy itself, it has a third equally indispensable feature of identifying certain ‘masks’ by their language as well as by other traits, so that given figures in the plays always use the same stylized dialect or linguistic register. (Pantalone speaks Venetian; the Dottore is Bolognese; Zani, Arlecchino and the other servants tend to speak Bergamask or generic Lombard; the Lovers use high-flown literary ‘Tuscan’. Only the braggart Capitano may vary between Spanish, Neapolitan, or the standard language. It is noteworthy that dialects which were popular in Venice but had too local a resonance, such as Pavano and Greghesco, disappear from a genre which was performed over a wider area.)

Cesare Molinari rightly points out (1985, pp. 9-13) that theatre of this kind cannot historically be restricted to professional performers, nor did the professionals confine themselves to performing commedia dell'arte (as is shown by the 1567 season in Mantua). Nevertheless there is an intimate link between the new breed of professional actor, who appeared in the mid-sixteenth century, and the improvisation of comic scenarios. The composition of most professional companies was dictated by the need to cover all the standard parts in a coventional comic plot, and stage names were almost always taken from commedia dell'arte masks. Most important of all, it is unlikely that improvisation method would have arisen at all in any context which was not professional and lower-class. On the one hand it is a technique which was probably forced into being by the illiteracy of some performers; on the other hand, as has been suggested by Roberto Tessari (1969), improvisation made it easy to multiply the number of different shows in the repertoire of a professional company. Scenarios could be composed more quickly than scripts, and could borrow material from a common stock in a less blatant way.

There has been much speculation on the origins of commedia dell'arte and of its individual masks. They have been related to various pre-existing figures of carnival and legend, sometimes going back to Roman times, and the performing structures have been traced by some to the Atellan farces of the pre-Christian era. For present purposes one point must be insisted on. All the ‘comic’ scenarios which have come down to us (and these are the vast majority, even though there are some which rank as ‘pastoral’ or ‘tragicomic’) base their plot schemes and characters on conventions established by commedia erudita between 1508 and the 1550s. The individual masks, whatever pre-existing images they may have drawn on, are chosen to fit the roles and functions normally demanded by such standard plots. A commedia dell'arte company categorized its members as Vecchi (‘Old Men’—initially Pantalone and the Dottore), Innamorati (‘Lovers’) and Zanni (‘Clowns’—low-life characters, usually in the role of servants): sometimes the women were listed separately, but it was clear that they in turn were either Innamorate (‘Lovers’, Heroines) or Fantesche (‘Maidservants’).⁶ This obviously fits the kind of framework which we have seen becoming standard in the less inventive scripted comedies, both in Venice and in Florence. The Innamorati were to be obstacled by the Vecchi, who would be moved by motives of avarice, of obstinacy, or of senile lust; the Zanni would try to help the lovers and play tricks on the old men, but would also be the butts of comic violence and disaster when their conspiracies went wrong. The braggart Captain—and this also reflects scripted sources—might take the role either of ‘lover’ or of ‘father’, depending on the chosen story line and the resources of the company. Happy endings were procured, as in ‘regular’ comedy, by a combination of successful trickery and the discovery of lost family relationships.

Commedia dell'arte was thus an improvised form of theatre spinning off from a scripted form, an adaptation of a gentlemanly cultural pastime to a format which was more commercially viable. If its masks and clichés are also to be found in undisciplined carnival culture, it is likely that they were transferred from the professional stage into that culture, rather than vice versa. The single undisputed case of the genre picking up a popular legendary figure is that of Arlecchino or Harlequin. The other ‘Zanni’ masks derived from simple caricatures of a low social class, the Bergamask peasant immigrant labourer. (Zani, or Zane, was a Lombard dialect reduction of the common name Giovanni.) Harlequin, exceptionally, was imported from France,⁷ where his name was that of a demonic elf, leader of the Wild Hunt, possibly the Erl-König of Germanic folk-lore. He was assimilated eventually, at least in his language, to the other Zani figures, but his patched costume and bestial black mask remained distinctive.⁸

Individual actors normally played the same ‘mask’ for their whole careers—this was a specialist trade, like that of other Arti or guilds. The use of the term ‘mask’ in this context is not always consistent from one scholar to another, so we had better define our terms. In one sense a mask is a facial disguise which needs no explanation, and facial masks were worn in commedia dell'arte by Vecchi and by Zanni, perhaps by some of the more grotesque Capitani. We can become familiar with them from the stock of contemporary visual material which is reproduced by major studies of the genre. In what we think of as the ‘classic’ form of the arte, young lovers, maidservants and some Capitani did not wear facial masks. It is better nevertheless to think of those roles as ‘masks’, rather than as ‘stock characters’ or some other term. All arte roles, whether facially masked or not, kept the same name, costume, language and other exterior characteristics from one play or scenario to the next. Leading ladies like Isabella Andreini were always ‘Isabella’ (or equivalent) on stage, always used the same language and delivery, always the same stage personality—they were fixed ‘mask’ roles in exactly the same way as Pantalone or Zani were. A familiar analogy would be the unchanging ‘masks’ adopted, without the use of a facial appendage, by actors in silent film comedy at the beginning of the twentieth century—Chaplin's ‘tramp’, Keaton's ‘wooden-face’, Harry Langdon's ‘baby-face’ are fixed and instantly recognizable figures, whatever social or psychological traits they may have to adopt for a particular story. Those cinema roles lived and died with a single performer, as no doubt did Isabella Andreini's ‘Isabella’, or Flamminio Scala's ‘Flavio’. The only difference with the facially masked roles was that the mask itself did not die, and so the role could be attempted by more than one performer. So there was a succession of Pantaloni, Zani, Dottori and Arlecchini over two centuries, with a constant dialogue and tension between the fixity of the mask and the personal genius of the actor.⁹ By contrast, each young lover and each pert maidservant had to be created afresh—though still affected by conventional parameters—in relation to a new star personality.

The wealth of surviving visual evidence (of which equivalents are totally lacking in relation to commedia erudita) has left a vivid impression of all these strange personalities. Scholars and enthusiasts feel that they know them well, and the fact that they can be visualized has given us a tendency to concentrate on the non-verbal aspects of arte performance—mime, gesture, acrobatics, and the mysterious qualities bestowed by the facial masks themselves. All of these were undoubtedly crucial; but it would be a serious error to underplay the verbal element in this kind of comedy. We have seen in the previous chapter how the obsession with dialect and caricatured language was a central reason for the ‘non-respectable’ comedians to separate themselves from more literary practices. Much of the fun of commedia dell'arte came from its performers saying, as well as doing, things which made people laugh. Molinari points out (1985, pp. 21-4) that the Dottore and the Capitano were essentially verbal masks, in that the ridiculous things which they said took priority over any contribution they made to the action. Both characters, indeed, came to life via bravura set speeches, in which the Dottore's idiotic parody of learning, or the Capitano's fantastic boasting, held the stage in their own right. We might also add that the Lovers too, whether their role is to be seen as a parody or not, made their presence felt principally by rhetorical exchanges.¹⁰ Pantalone and the myriad versions of Zani and Arlecchino may have injected more physical energy and clowning into the action, but they too, with their caricatured dialects and deliberate stupidities, used words as much as they used their bodies.

It is legitimate therefore to suppose that there must have been some interplay, if also some contrast, between the verbal content and style of commedia dell'arte and those of ‘regular’ Italian comedy. If improvised theatre grew out of a scripted genre, then it must have taken some of that genre's dramaturgy with it, on the small scale as well as the large. Later, when the arte was established and popular, writers of comic scripts could easily have been influenced, consciously or unconsciously, by what they saw and heard the professionals doing. The relationship between the two types of practice is bound to be complex, and it will involve (on both sides) rejecting as well as accepting influences; but some relationship has to exist, and sometimes it may still be detectable in the evidence which we possess.

It is the question of evidence, of course, which is the most debatable. From a purist point of view, there can be no commedia dell'arte text on which we can base any arguments. Since the actors improvised, whatever that may turn out to mean, from an outline scenario, they used no written script. Whenever we find a written script which looks as if it might be informative, then by definition it is no longer improvised and therefore no longer commedia dell'arte. This logical dilemma makes us rely, to a degree which is in theory dangerous, on personal judgement, even on hunch. In particular we have to decide empirically, not logically, what can rank in any dramatic script as ‘evidence’ of non-scripted performing practice; and that decision will depend largely on our having perceived features which we have decided, in advance, constitute the ‘evidence’ which we are seeking. The argument is circular, but unavoidable: we can comfort ourselves with the observation that most of the judgements we have to make in real life, as opposed to scholarship, are taken on the same unsatisfactory a priori basis.

In practice we have already taken some of the relevant decisions in the previous chapter, in quoting certain structures in written Venetian comedy as foreshadowing what was to develop into improvised theatre. It will be similarly necessary to use as evidence (relying heavily on the circular argument) later comic scripts which, in our judgement, show signs of being influenced by improvisation, and which therefore may inform us what detailed influence there might be. The need to get help wherever we can find it will force us to make use of material which falls well outside our usual chronological parameters, well into the seventeenth century. There are two bodies of plays in particular which have proven links with commedia dell'arte, and where it seems plausible to detect elements in the dramaturgy which come from improvisation. One is made up of the works of Molière, who is known to have had regular contact with Italian arte performers in France.¹¹ The second is a group of comedies now referred to by critics as ‘Commedia Ridicolosa’.¹² These are fully scripted plays, written by amateurs for amateur performance in or near Rome, and published mostly between 1605 and 1630. The most popular authors (that is, the most often reprinted) are Giovanni Briccio (1579-1645) and Virgilio Verucci (dates unknown). The link between ‘Commedia Ridicolosa’ and commedia dell'arte comes on a secondary level from a proliferation in these scripts of regional dialects (one of Verucci's plays is actually entitled Li diversi linguaggi¹³—‘Various Languages’, first published 1609); and on a primary level because the main characters in the plays are Pantalone, Zani, Arlecchino, Dottor Graziano, Franceschina, various Capitani and young lovers, plus other stereotypes of more ephemeral local interest. The accepted view of these plays is that their authors and performers wanted to reproduce something like commedia dell'arte in private performance, but lacked the ability to improvise with confidence and so needed the support of a full script. Therefore, we can argue, at least a part of their writing might reproduce the style and atmosphere of an improvised performance.

It is with reference to this wide and disparate range of evidence that we must propose a theory of what ‘improvisation’ meant in practice, and of what characteristics it may have tended to impose on the structure of a comic speech or scene.

IMPROVISATION AND MODULAR STRUCTURE

In his most recent study (1984), Roberto Tessari explores the tension which existed among arte actors, before and after 1600, between the more respectable and less respectable ends of the profession. The premise is that commedia dell'arte grew out of a socially unacceptable form of theatre, and that although by 1600 the actors were seeking and gaining a new status, they were still aware of, and nervous about, their historical origins.¹⁴ The new status came from their having adopted the forms of respectable humanist comedy; however, the method which the professionals applied to this material, the method in particular of improvising rather than learning from a script, came from elsewhere. We can only conjecture that these practices already existed, and must have been adapted from such professional actors as were already performing: individual buffoons, and itinerant family companies who acted in the open and who had no cultural or social status at all.

It is likely that such performers were mostly illiterate, and that their approach to creating ‘scripts’ and memorizing them for performance would be different from the methods adopted by anyone who could read and write. In the first place, like comedians of all times and places, they would want to build up a mental store of single jokes, ranging from one-liners through question-and-answer routines to quite complex verbal and physical sequences. These would be created, learned and rehearsed piecemeal, and then used wherever they would fit, with the narrative line being adapted to the joke, if necessary, rather than vice versa. Such jokes would be the ancestors of what in arte jargon were eventually called lazzi. However, as well as a stock of autonomous gags, these actors would need a technique for creating and memorizing material which actually led somewhere, dialogue which was capable of building narrative and therefore of contributing to the plot. We are cut off, now in our society, from the mental processes of the creative intelligent illiterate—nevertheless, it is not impossible to hypothesize at least one of the mental techniques which could have been employed. Functional dialogues would be easier to memorize if they were built out of small manageable units, not so very different from the autonomous so-called lazzi. Each stage of a dialogued exchange would be identified in the mind by some label—the information it conveyed, or the joke it explored. In memorizing and performing the sequence, the stress would lie on conveying the information and/or getting the laugh, rather than on repeating the words identically every time.

This hypothesis relies heavily on the circular argument, because it was formed in response to the fact that certain relevant dramatic texts fall into a characteristic pattern—one whereby dialogue is built out of short units, many of which are interchangeable, or removable, or indeed recyclable into a different narrative context. Some examples have already been quoted in the previous chapter from Aretino, Dolce and Parabosco, and it would have been possible to add others from Calmo and Giancarli. Later, examples of the same structure are plentiful in ‘Commedia Ridicolosa’ and in Molière; but our starting point, and our basic illustration of the concept, is a single precious text from the period which concerns us.

Vito Pandolfi, in his indispensable six-volume collection of material published between 1957 and 1961, offers an anthology of surviving textual material relevant to the arte. Most of his examples are not dramatic texts as such, but spin-off material in the form of songs, verse dialogues, and pamphlet compositions often also in verse. This is hugely informative about the way in which Pantalone, Graziano and Zani made people laugh, but it tells us little about how they structured their jokes for improvisation on stage. But there is one exception: a dialogue in prose between the Venetian Magnifico and his servant Zani (Dialogo de un Magnifico e Zani Bergamasco),¹⁵ which embodies the central master-servant confrontation, placed firmly in the Republic of Venice, which many scholars see as the core of commedia dell'arte. The text has an unusual ring of live performance about it, to the extent of making us wonder if it is a transcription of a scene actually staged, or a guide for new performers learning the trade. It could comfortably function as a scene—or as two scenes interrupted by other material—in a scripted play or in an improvised scenario.

The plot is simple and typical. The elderly lustful Venetian Magnifico wants to make contact with a courtesan. He sends his servant, country bumpkin Zani from Bergamo, into the lady's house with a message and a poem. After an interval, Zani comes out again, having had such a good time that he totally forgot to do what he was sent for. There are thus two distinct routines, both familiar from scripted comedy: one in which the old man reveals to his servant that he is in love, and is mocked for his pains; and a second in which the servant confesses his incompetence and is punished for it. The first ‘scene’ is further sub-divisible into shorter routines, of which the following are a sample:

(A)

MAGNIFICO:
[Introductory monologue, then …] Zani!
ZANI:
Yes. sir. [Probably at the window]
MAGNIFICO:
Come out a moment.
ZANI:
What can I do for you, boss?
MAGNIFICO:
Come out here a moment, my dear chap.
ZANI:
Me?
MAGNIFICO:
Thee.
ZANI:
You want me to come to thee?
MAGNIFICO:
Aye, thee, you donkey, get a move on.
ZANI:
At your service, as you see. Shall I wear my hat?
MAGNIFICO:
To hell with your hat, put your cap on.
ZANI:
Yes siree. Er … Pardon me …
MAGNIFICO:
Now what?
ZANI:
You want me to come out?
MAGNIFICO:
Yes, you.
ZANI:
Me myself in person?
MAGNIFICO:
YES, for God's sake, come out!
ZANI:
I'm coming, I've arrived, I'm here, what can I do for your honour?

(B)

MAGNIFICO:
I want you to know that I am in love.
ZANI:
In love?
MAGNIFICO:
Yes, I am in love.
ZANI:
You mean you are in love?
MAGNIFICO:
Yes, idiot, don't you understand?
ZANI:
You really mean it?
MAGNIFICO:
Yes, I really mean it.
ZANI:
You are in love?
MAGNIFICO:
Yes, I am in love.
ZANI:
(Collapses with laughter)
MAGNIFICO:
What are you laughing at, you clodhopper?
ZANI:
I thought you'd made a joke. You said you were in love.
MAGNIFICO:
But I am in love. What's the matter, don't I look like a man with normal drives?
ZANI:
Driving mules, more likely.

(C)

ZANI:
And you're in love with her?
MAGNIFICO:
She's the one.
ZANI:
Then you've had it, boss, you're as good as dead.
MAGNIFICO:
Dead?
ZANI:
Dead, boss, you're dead.
MAGNIFICO:
But I'm perfectly alive, you dolt, what's all this about?
ZANI:
You're dead, boss.
MAGNIFICO:
Why am I dead?
ZANI:
Four brothers, real hotheads, they'd clobber you as soon as look at you.

The three ‘sections’ chosen above do not, of course, appear as separate in the text. However, they are easy to isolate and consider separately. (In fact, (B) follows almost immediately after (A), whereas (C) comes after two or three intervening sections.) They are autonomous units of dialogue, each reaching a simple conclusion (which may also involve a joke), and each is potentially usable in other situations. They can be seen as a series of beads threaded on a string, or as a set of blocks which together will build a wall—in either analogy, the structure of the scene which they compose can be described as ‘modular’.

In addition, the single units of the modular structure are in many cases ‘elastic’: they can be made longer or shorter, in improvisation, without losing their essential point. If the opening suspense about whether Zani is going to come on stage or not falls flat, then the actors can skip some of what is here written down and get on to the next unit: the same is true of the ‘I am in love’ sequence; while the section about the Magnifico being ‘dead’ could be lengthened with more repetitions, if the actors have struck the right rhythm and the audience is happy. What the units have in common is that their conclusion, or punch line, is never in doubt, so the actors cannot get lost—all they need to do is to identify each sequence by its climax and get them in the right order. For a smooth performance there has to be a previously agreed cue line or gesture, which brings the sequence to an end. So, in our section (A) above, when the Magnifico says ‘Yes, for God's sake …’ with the agreed intonation or emphasis, the actor Zani accepts that he must come out of the house, and forget any further delaying variations. In the ‘You're dead’ sequence, it is perhaps Zani who is in control—he is the one who has to judge when to give the answer about the four brothers, and move on to the next unit. It was that kind of professional discipline, subordinating the single effect to the overall flow of the spectacle, which was later recognized as distinguishing a good company from a bad one. If either the Magnifico or Zani is selfish, and ignores the agreed signals to move forward (or exit, or enter), then the teamwork is spoiled.

By the nature of things, ‘elastic gags’ or modular units are often suspense structures built up by sheer repetition (as we also saw in extracts from Aretino and Dolce). This is obviously the easiest shape to use, if the length is going to be varied. It is a relatively simple business for the actor to go on saying more or less the same thing until some signal tells him to stop. In fact any structure involving suspense and a postponed conclusion, repetitive or not, poses few problems. As long as the prearranged climax is waiting as a safety net, the actors can perform almost any verbal acrobatics without losing their balance. This process reaches its peak in the second scene of our dialogue, when Zani comes back from the courtesan's house and keeps his master poised on the edge of apoplexy waiting for his news:

ZANI:
Good news, boss! I've been there.
MAGNIFICO:
You've been where?
ZANI:
To see your girlfriend. Good news!
MAGNIFICO:
Really?
ZANI:
Really, boss.
MAGNIFICO:
My dear old chap, come on now, give me some consolation, give me some relief!
ZANI:
Boss, I was in there talking to her for ages.
MAGNIFICO:
Splendid! What did she say about me?
ZANI:
She's so polite, so accommodating, so friendly.
MAGNIFICO:
Yes indeed, she's got all the graces. What did she say about me?
ZANI:
She gave me an enormous hunk of her cheese.
MAGNIFICO:
Get to the point man, tell me what she thought of the sonnet, and what her answer was.
ZANI:
She gave me some fresh, white bread.
MAGNIFICO:
Do you want me to burst? You can tell me about all those things later—put me out of my misery, tell me about the sonnet.
ZANI:
Yes boss, just a minute boss, good news! She wanted it, you know sir, she really wanted it.
MAGNIFICO:
Wanted what?
ZANI:
She wanted it from me.
MAGNIFICO:
WHAT?
ZANI:
She took hold of my hand, and she wanted me to touch her on her rosette.
MAGNIFICO:
On her what? Her rosette?
ZANI:
Oh yes, boss … And she wanted me to stick two buttons on it.
MAGNIFICO:
What the hell is all this? What buttons, what rosette?
ZANI:
Yes boss, you see, the rosette on the front of her bonnet, here … She wanted me to give her two of my buttons, to stick on her rosette. It was a great favour.
MAGNIFICO:
God rot you, you great buffoon, will you stop this gibbering and tell me what she said about me?
ZANI:
Oh boss, she was smiling, she was happy, she was so kind to me, she said if I go back again she'll give me some cake.
MAGNIFICO:
I don't think I can stand this much longer—I'm going to die. WHAT DID SHE SAY ABOUT MY MESSAGE?

And finally he does get an answer—but it is that Zani forgot to give the message at all, and used the sonnet to wrap up some fried fish. The scene ends in tears, with a beating. There is no verbal repetition here (except for the irritating phrase ‘Good news!’ to be inserted at frequent intervals), but an inventive Zani, on the night, could no doubt think up some more dreamily suggestive double-entendres and spin out the comic delay for as long as the audience continued to laugh.

The whole sketch, then, appears as a flexible succession of ‘elastic gags’, or modular units of dialogue. Our proposal is that such modular components are a sign of improvisation technique, of a mode of performance in which an actor's existing repertoire of jokes, long and short, can be adapted and inserted into any plot with which they do not actually clash. Such techniques could have been developed first of all by illiterate performers, who needed them to help build their mental catalogue of material: they were then applied increasingly, especially in Venice, to the new ‘erudite’ material which was thus captured for professional use. When the acting profession ceased to be illiterate, as it did very rapidly around this time, then the mental memory bank was replaced by the personal libri generici (commonplace books), written collections of useful and transposable material which we know actors kept right down to the time of Goldoni.

If all this is true—and it is by no means a revolutionary suggestion, in the light of what is already known and documented—‘improvisation’ did not mean that an arte performer invented his part from scratch every night, but rather that actors, singly and in teams, were constantly drawing on an accumulated stock of existing repertoire. The plots which they used were in any case stereotyped and repetitive, constantly offering similar situations. Creative adlibbing could never have been ruled out—the genre would not have been such a phenomenal success without it—but when it happened it would be like a jazz musician playing variations on a known tune,¹⁶ securely held by an existing framework and always moving towards a known conclusion or comic climax. And the structure of autonomous repertoire ‘melodies’ would not have been restricted to large-scale lazzi, monologues and long speeches, but would operate also in fast-moving dialogue, even in exchanges which advanced the plot rather than merely embroidering it.

Although we found some examples of modular structure in written comedy before 1555, it seems to be relatively rare; and this underlines a divergence of taste and aspirations as between ‘literary’ and ‘theatrical’ dramaturgy. Most of what we have quoted from the anonymous Dialogo is unexciting to read on the page. Writers with a literary training avoided circularity and repetition, which to them appeared as literary faults. (As we noted, when Florentine dramatists found such playful structures in Plautus,¹⁷ they tended to exclude those scenes from their ‘imitations’.) They wanted their comedies to perform well, but also to read well. The professional buffoon did not expect his material ever to be read, and by adopting improvisation even when he was literate he ensured that it would never be. He knew that repetitive and even silly material can work beautifully in the third dimension of live performance.

The tracking down of modular structures and ‘elastic gags’, in those written scripts which might be affected by the example of arte improvisation, is a task which has barely begun. Surveys of the appearance of such a structure in ‘Commedia Ridicolosa’ and in Molière are nevertheless encouraging. They even make it possible to point to particular types of scene, conventional and frequent in the ‘regular’ comic plot as developed between 1550 and 1600, in which the modular structure can be especially useful.

The anonymous Dialogo shows that one of the most obvious purposes of the modular game can be to cause delay and create suspense. The event which is delayed can be some major comic climax, like a beating or the delivery of unwelcome news; or the most banal thing imaginable, like the opening of a door. An unimportant action is made important simply because it is put off for so long: an event which is not funny in itself, when it happens, can be made hilarious because of a build-up of time during which it does not happen. (It is perceptions like these which a purely literary dramatist, with no performing experience, is slow to acquire.) In Act i, Scene 4 of Giovanni Briccio's Pantalone imbertonao (‘Pantalone Besotted’) of 1617, the delay in getting a servant to come out of a house is taken to greater lengths. Young master Tiburzio wants Zani to come down and let him in. But Zani is having a meal, and puts off moving for ‘just one more mouthful … just one more glass of wine …’, in a repetitive elastic sequence. Tiburzio naturally gets furious, and picks up a stick. But then Zani does not want to come out because he is afraid of being beaten, and there is a new elastic delaying sequence in which Zani says ‘You'll hit me’ and Tiburzio says ‘No I won't.’ When Zani finally does open the door, he does of course get beaten, as the audience anticipated all along. Much later, in Molière's L'Ecole des femmes (‘School for Wives’), Act i, Scene 2, there are two servants, Alain and Georgette, to open the door to their master Arnolphe. To start with neither of them can be bothered, and there are elastic repetitions (‘You go.’—‘No, you go’) accompanied by fatuous excuses. Then Arnolphe threatens to starve them as a punishment, so they start falling over each other to open the door, and the contest between them to avoid blame causes more delays. The simple interposition of a door between characters seems to provide endless opportunities for such games.

A different opportunity is given by the reading of a document on stage. This can be done simply to invite repetitive comments from a listener, as in Act ii, Scene 1 of Molière's L'Avare (‘The miser’). In ‘Commedia Ridicolosa’, there are scenes in which a dowry is laboriously read through by the Dottore, acting as marriage-broker. This character's Bolognese dialect mixed with bad Latin produced complete gibberish: so, as he pieced his way painfully through reading the dowry contract, each impossible item of property had to be translated back into its proper sense by some other character on stage. So (to fudge up English equivalents, but with some Italian originals in mind), the meaningless phrase

One lousy farthing for an awkward haddock with pox

turns out to mean

One house and garden with an orchard, paddock and copse;

while

Nick wet the bed milking Persians for Mother Eileen

is really

Six sets of red silken curtains and another in green.

This sort of thing is harder to invent on the spot, and the distortions might have to be memorized; but if the ‘real’ meanings were written out on the document used on stage, the Dottore could exercise his judgement as to how many to use on a given night. Versions of this scene, in written form, appear in Briccio's La dispettosa moglie (‘The Bad-Tempered Wife’), ii, 3; and in Pantalone imbertonao, iii, 4.

Another kind of dialogue which invites repetitive modular structure is one which works through a list of alternatives, as long or as short as necessary. In Briccio and Verucci there are leisurely reviews on subjects like what to do next (Pantalone imbertonao, iv, 4); the virtues which can be detected in a prospective bride (La dispettosa moglie, ii, 4); and alternative places where Zani might go to hide (Li diversi linguaggi, v, 1). In Aretino's Talanta, in a scene probably written for professionals, we saw Tinca giving a list, in both words and gestures, of the ways in which he could frighten off his enemies, and such a structure could well be the basis for a large part of the repertoire of any braggart Capitano. But the most typical list of all is a list of people; especially a list of suitors for the hand of a son or daughter, where a parent is broaching the subject of marriage. In Briccio's La dispettosa moglie the same structure is used in two successive scenes (1, 2 and 3): the father works through a list of possible candidates, the son or daughter finds a reason for rejecting each in turn, until the suitor already favoured by the young person is mentioned and modestly accepted. Or, in a similar context, an argument can develop between parent and offspring over a marriage proposal, which can be reduced to the simplest repetitive structure of all—the one where one party is saying ‘Yes’ and the other is saying ‘No’, ad nauseam, until a third character breaks in to stop them. Act iii, Scene 1 of Verucci's Li diversi linguaggi is an expanded form of such a confrontation, which I have analysed in detail elsewhere.¹⁸ There are a number of successive ‘units’ in it, but all of them are elastic; and for good measure various of them recur, as the same routines but with different words, in separate scenes of Molière's L'Avare.

The confrontation between parent and child over a proposed marriage is a standard conflict in plots derived from commedia erudita, and a good example of the way in which professionals would want to milk a situation for its maximum number of laughs. In most Florentine and Venetian comedies where there is rivalry between father and son for the same girl, that situation is presented as antefact in an early expository scene, without much dramatic exploitation. In later comic tradition it is seen to provide an important opportunity for conflict on stage. In a scenario, it might be very simply summarized as ‘Pantalone speaks to his son Flavio about marriage, and reveals that he wants to marry Isabella.’ In practice, the modular structure of an improvised dialogue is going to delay the revelation as long as possible, and get some fun out of misunderstandings on the way. Father and son might spend some time discussing the desirability of marriage in general, appearing to be in full concord; they might then agree on the desirable qualities of Isabella in particular, with Flavio unable to believe his luck; only then would Pantalone drop the bombshell and reveal that he wants this girl for himself. That, at least, is the way it is done by Molière in Act i of L'Avare, where the father and son are Harpagon and Cléante. And since half the amusement of such a build-up comes from the audience's guessing in advance that Flavio is misunderstanding the situation, one can speculate that by Molière's time the routine was very familiar. We have not been explicitly told, at that point, that Harpagon wants to marry Marianne, but we will enjoy the scene more if, on the strength of many previous comedies, we can guess that this might be the case.

The appearance of elastic routines in Molière generally—and they are extremely numerous—should not be taken to indicate that his scripts are provisional, or that he intended his actors to ad-lib them in performance. Rather, he must have observed a particular rhythmic structure which was adopted by improvising actors, and seen that it had a theatrical validity in its own right which could be turned back into a written script. Some of his most penetrating comic scenes use the structure, such as the famous repetition of ‘Le pauvre homme!’ in Tartuffe; but his judgement is so perfect that any attempt to vary what he has written would spoil it completely. The most concentrated example of his using elastic gag structures is the whole first act of Le Médecin malgré lui, where performing experience shows that there is not a single word too many or too few.

Modular structure tends to accentuate the self-conscious game-playing element of comic theatre, as well as having the practical advantage of accumulating repertoire for the individual actor and making things easier for his hard-pressed memory. Having identified it as a technique, we must not allow ourselves to be mesmerized by it. The elastic gag cannot have been the only characteristic of the unwritten dramaturgy of improvised comedy. Nevertheless, when it is found in a written script, it is a useful sign of the presence or influence of professional arte practice; and in Italy it helps to highlight the grudging interplay between improvised commedia dell'arte and its parent, written commedia erudita.

LARGER UNITS OF IMPROVISATION

The earliest type of input into ‘regular’ scripted comedy from the performing routines of professional buffoons seems to have been in the form of the large-scale monologue. An essential part of the skill of a buffoon or giullare had always been that of holding the attention of an audience without help, often splitting himself into multiple characters or voices in order to do so. Ruzante, Calmo and Giancarli all tended to isolate such moments in which they individually, as actor-dramatists, built a confidential relationship with their public. In their more lengthy compositions, they allowed some of their colleagues to do the same, thus contributing to a tendency inherent in all comic drama, whereby the spectacle fragments into individual bravura numbers and the audience savours them one by one.

Every mask in the canon would have been capable of taking the stage alone in this way; and there is still work to be done in tracing not only what verbal formats would be typical of each, but also the ways in which they may have influenced written comic scripts. In the years before the genre was fully established, the plays of Ruzante and Calmo contain some of this material in embryo, with Ruzante's monologues pointing in the direction of foolish one-man numbers for a Zani or an Arlecchino, and Calmo laying the foundations for monologues by Pantalone. There are also foreshadowings of the braggart Capitano mask in various plays, both Venetian and Sienese, which caricature the figure of the Spanish soldier; and Pedants in Aretino, in Belo and in Gli ingannati point—perhaps more tentatively—in the direction of the mask which will be Dottor Graziano. Capitano and Dottore in particular were identified by characteristic types of monologue or long set speech. In Pandolfi's anthology, jokes and themes which could also be used on stage are preserved in print in semi-dramatized songs, verses and pamphlets: most such compositions (though not all) tend to work in terms of longer ‘speeches’ rather than rapid dialogue. The so-called ‘verbal masks’—Dottore, Capitano and Lovers—would have been especially prone to accumulating repertoires of long items, which may have needed memorizing more carefully because the verbal details were the essence of the effect sought.

The Capitano was a role much in demand during the ‘classic’ period of the genre, but had disappeared entirely by the eighteenth century. Pandolfi's material shows that his boasting could fall into two distinct categories, each no doubt preferred by different performers of this family of masks.¹⁹ They are typified by two contrasting poems in ottava rima, each of which encapsulates for a reader the essence of one version of the theatrical role. The one by ‘Captain Headsplitter and Arrowspitter’, written (and performed?) by Antonio Pardi from Lucca,²⁰ goes for straightforward impossible hyperbole: ‘I break, smash, shatter, fracture and dismember anyone who steps in my path … Many times have I broken the head of Cerberus, and routed the assembled ranks of Tartarus. Wherever I walk I can put Pluto and Satan to flight … And if my name is heard [in the heavens], then Hercules and Mars shit in their breeches.’ Long rantings of this sort seem to have been remarkably popular, even the endless allusions to Classical mythology and abstract personified figures such as Death and Tumult, though one must always wonder whether the material was pushed culturally up market just for the printed page. This is very much the style of the famous Francesco Andreini (1548-1624), in his prose Bravure (‘Boasts’) of 1607,²¹ which to judge by their introduction are actuely anxious about whether their material will be acceptable to literate society, and to ecclesiastical censors. An alternative Capitano, as exemplified by the ‘Earthshaker’ of Giulio Cesare Croce,²² might go to the other extreme and describe the most banal or ridiculous exploits in inflated language, as though they were superhuman: ‘I decimated a plateful of beans single-handed … And a Cricket once, who was singing unscathed in the forest, I assailed with such fierce and horrendous cries that he fled into his hole and I never saw him again.’ This ‘Seven [Flies] At A Blow’ approach, as well as having roots in folk-tale, is closer to the one adopted by Pietro Aretino's Capitan Tinca, in La Talanta of 1542:

TINCA:
Do you know that at the battle of Cerignola, which lasted until after dark with one man-at-arms dead and two wounded, I was the man who fetched the fire which lit the torch carried by the man who stepped between the combatants, quelled each side with a glance, and said: ‘Gentlemen, you have done enough for today’?

(La Talanta, Act iii, Scene 12)

The fact that Tinca needs his parasite Branca as a ‘feed’ typifies the way in which a Capitano usually worked, with a companion always on stage to flatter him, provoke yet more boastings, and ironize for the audience: Andreini's published Bravure may differ considerably from what he did on stage, but they do retain the essential dialogue between ‘Captain Terror’ and his servant Trappola (‘Trap’). The repertoire must have consisted of quite long complex sequences, largely memorized by the Capitano himself, but with practised interjections, following predictable patterns, from the servant or parasite.

The repertoire of the comic Doctor, on the other hand, mostly took the form of monologue—at least it could have been prepared and memorized as monologue, even if other characters were going to listen to it, because his speciality was to spin out any conversation to interminable lengths of verbiage, and over-ride all attempts to interrupt. So, at least, we gather from a late appearance of the figure in one of Molière's earliest surviving farces, La Jalousie du Barbouillé (‘The Jealousy of Barbouillé’). The Doctor makes three appearances in this short piece, two of them lengthy and none of them in any way useful to the action or to the other characters—the whole basis of his comedy is that his endless drivel prevents everyone else from getting on with their business. In Scene 2, Barbouillé (‘Doughface’) makes the mistake of consulting this learned man about his marital problems, and has to listen to a tirade demonstrating that the Doctor is a doctor ten times over. The fact that Barbouillé keeps trying to interrupt turns what is basically a monologue into a modular delaying routine:

DOCTOR:
Firstly: just as the unit is the basis, the foundation and the origin of all numbers, so I am the first among doctors, the Doctor of doctors. Secondly: since there are two faculties necessary to perfect knowledge of all things, namely sense and understanding; and since I am all sense and all understanding, thus I am a doctor twice over.
BARBOUILLE:
Absolutely. The thing is …
DOCTOR:
Thirdly: since the number three is the number of perfection, as Aristotle says; and since I and all my writings are perfect, so I am a doctor three times over.
BARBOUILLE:
Yes. Well now, Doctor …
DOCTOR:
Fourthly, since philosophy is divided into four parts …

(La Jalousie du Barbouillé, Scene 2)

Barbouillé has no choice but to sit back and wait until number ten has been reached. Such ill-founded pretentiousness was always a characteristic of the Italian mask Dottor Graziano. Pedants in written comedy have the same characteristic, but it is not always expressed in the same repetitive games. However, one early example which does offer pre-echoes of Molière's Docteur is the pedant Aristarco in Bernardino Pino's otherwise untheatrical Gli ingiusti sdegni of 1553. What does not appear in scripted Italian comedy, and is lost both in Molière's French and in any conceivable English translation, is the barrage of twisted language which characterized the professional mask. The Dottore spoke Bolognese, a dialect which Italians from elsewhere tend to find amusing in its own right. To this accent he added streams of bad Latin (for the educated), or better still (for the uneducated, or for everybody) an endless flow of distortions and malapropisms which became a comic code in its own right, and was reputed in the mouths of the best performers to make audiences literally sick with laughter. A ‘Gratianesque Dictionary’²³ compiled in manuscript for the help of aspiring performers gives a list of the distortions to which certain words should always be subject, most of them untranslatable in any literal sense. Parer (‘opinion’) should become sparvier (‘sparrow-hawk’); Capitano becomes Decapidan (a non-word from decapitare, ‘to behead’); ordinar (‘to arrange’) becomes, naturally, orinar (‘to urinate’); letto (‘bed’) becomes letame (‘manure’); and we know from earlier written texts including Ruzante that the academic discipline Medicina was regularly twisted to Merdesina (based on merda, ‘shit’), and Latino to variants of latrina. Tastes were more robust and simple in those days, but one must not underestimate how laughter can come from accumulation—if every second or third word in a sentence was distorted in this way, the effect would have been entirely different from that of each piece of word-play on its own. Molinari (1985, p. 22) quotes a snatch of a letter in Gratianesco gibberish from an early Dottore, Ludovico de' Bianchi, to the Grand Duke of Tuscany: every single word is twisted into something else, and it takes considerable ingenuity to get any meaning out of it at all. No doubt audiences would build up an acquaintance with this code, and eventually congratulate themselves on seeing through it to the banalities which lay underneath.

How banal the content could be is shown by another spin-off composition in verse: the One Hundred and Fifteen Conclusions of the Archperfect Doctor Gratiano,²⁴ the first stanza of which might go like this in English:

1 A rose, when blooming, tends to smell all right;
2 And any man who's walking can't be dead;
3 A man who's always wrong is never right;
4 A ship at sea is far from home and bed;
5 If peace and quiet annoy you, start a fight;
6 And if you won't walk slow, walk fast instead;
7 A childless woman isn't often fecund;
8 You'll always miss the first place if you're second.

(Just eight ‘conclusions’ out of one hundred and fifteen.) But the core of the Dottore's role was a prepared repertoire of long implacable speeches, full of tongue-twisting virtuosity. The enforced pause in the dramatic action would make them almost into spoken arias, with the gibberish taking the place of melody. Any similar structure in a written comic script can probably be traced back to this mask.

The material which still leaves scholars feeling very hesitant is that used by the Lovers, a class of mask which included that brand new theatre phenomenon, the Leading Lady or (in every sense) Prima Donna. All the surviving evidence suggests that the repertoire and style used by the Lovers was unremittingly literary. In the previous chapter we quoted brief examples from Ruzante and from Parabosco of the kind of preciosity which their dialogue could involve, and those two examples highlight a major difficulty in modern interpretation. In Ruzante's Vaccaria, the tortuous verbal agonies of Flavio and Florinetta were brought down to earth by immediate contrast with the vulgar dialect of Vezzo and Truffo. In the passage quoted from Parabosco's Hermafrodito, on the other hand, there was no reason to suppose that Zerbino and Furnia were not to be taken seriously. Which of these two tendencies, or what balance between them, prevailed in commedia dell'arte? Was the verbosity laughed at, swooned over, appreciated as serious rhetoric, or made a sly cover for the sexuality contained in the plot? There is some evidence which points to each of these, and perhaps ultimately the answer is complex rather than simple.

Pandolfi in his anthology makes use of two principal sources, fragments put together in the early seventeenth century by the working actors Domenico Bruni of the Confidenti company,²⁵ and Isabella Andreini (1562-1604)²⁶ of the Gelosi, wife of Francesco (‘Capitano Spavento’). The dialogues of Isabella, edited for publication by her widower, are more formal and literary, but both collections could relate to genuine items of repertoire. Bruni's unpublished dialogues are more prepared to admit an element of sexuality, even bawdry, into the relationship between hero and heroine, which corresponds better to the content of many surviving scenarios. Isabella's edited works are more decorous: this reflects the respectful adulation which she received in her lifetime, her admirers going to great lengths to stress her virtue and high moral tone. Both collections are good evidence for the view that actors learned and prepared material to cover stock situations, which would regularly recur in standard comic plots. Bruni's fifty-one items include the following titles:

A despised lover, in hatred of his rival.
The beginnings of love.
Whether beauty and grace are the same thing.
Lover's compliments at the window, on leaving his mistress.
A seducing bawd, on constancy in love.
Superiority of the soldier or the man of letters.
Parting.
Woman offended with her husband.
Man offended with his wife.

Isabella's twenty-seven items include:

Amorous debate on the dignity of lovers.
Debate on the passions of hatred and love.
Debate on those who have died for love.
Debate on the fever of love.
Debate on how to fall out of love.
On loving another more than oneself.
On jealousy.

That the role of the Lovers was not always approached seriously is plain from the situations in which scenarios often involve them—elopements, sexual escapades, pregnancies, and mad scenes for both heroes and heroines. One of Isabella's own dialogues, the ‘Amorous debate on pretending to love one woman while loving another’, between Valerio and Fedra, ends with an extended ‘mad speech’ for the hero Valerio, which is clearly meant to cause laughter with its nonsense and non sequiturs in the same way as was seen in ‘drunk scenes’ and ‘sleep-walking scenes’ in Calmo and Giancarli:

VALERIO:
Now indeed night has fallen: now indeed the darkness of her cruelty invades and obscures my mind: now indeed I am blinded entirely! Ah! most cruel woman, you yourself said it, you yourself foretold it, that I should become all of a sudden the wisest man in the world, and should take over the lordship of the universe: and she spoke true! Hey, you, whoever you are! Bring me my royal robe, my sceptre and my crown; and get these rags off me, they are unworthy to clothe a King, an Emperor, a Monarch such as myself. [Possibly he should remove or disarrange his clothes at this point?] Come now, dress me quickly! There: now that I am royally dressed, bring in all my courtiers, summon my counsellors, because I wish them to advise me about my love, and about buttermilk curds. You, young gentleman, who gave you leave to carry that pistol loaded with Malmsey wine, in defiance of our edict? Take it away, and roast it, because I want to eat it stewed, casseroled and fried in the pan. Musicians! Play that dance tune which starts with the first of April and finishes with the end of May, because I wish to give a lunch for certain friends of mine who are fond of composing verses dressed in petticoats. Who would have thought that monkey knew so much about affairs of state? And yet it does; but if Aristotle hadn't drummed his Politics into its head, that great bell of justice would never stop ringing, and the sparrows would eat up all the millet in Lodi. Calm down, madam, there'll be plenty left for you …(27)

This continues for more than twice the amount quoted, with some of the nonsense perhaps containing topical allusions, and it ends with a demented snatch of disconnected folk-song worthy of Ophelia. Isabella herself was famous for her own mad scene in the scenario La pazzia d'Isabella (‘Isabella's Madness’), of which we have an account from 1589: whether she was as fey and ‘comic’ as this, or whether she went for more dignity and pathos, we cannot tell. But on the technical side, we have to conclude that most of this lengthy monologue would be learned by heart, perhaps split into independent sections, certainly with room for variations to suit each performance. In 1589, on an occasion to which we shall return, she was acting for a Granducal wedding where the bride was a French princess: so in her mad scene ‘among other things she began to speak French, and to sing various French ditties, which gave inexpressible pleasure to the Most Serene bride’.²⁸

The role in the Lovers' repertoire of solo numbers like this is not difficult to imagine. The dialogues in Pandolfi are harder to fit conceptually into the framework of ‘improvisation’, and Isabella's examples all seem to demand that they be simply learned by heart. They do not even have the embryonically ‘modular’ structure which we claimed to identify in the exchange from Parabosco's Hermafrodito quoted in the previous chapter. Nevertheless, that scene from a written play shows that even lovers' dialogues might be composed of smaller interchangeable repertoire items. Many of these would be what Elizabethan and Jacobean England called ‘conceits’, figures half-way between the poetic image and the legal argument, with a core of abstract content which could be divorced from a precise verbal formulation. Other units, as the Parabosco scene suggested, might have a more rhythmic character. Much later than our period, in 1699, a compendium on the art of improvised performance by Andrea Perucci contained a series of strange dialogues ‘in parallel’ between lovers, in which the partners declaim their way through a complex dispute and change of heart, each always expressing exactly the same set of emotions, but never in the same words.

HE:
Go! …
SHE:
Disappear! …
HE:
… from my eyes …
SHE:
… from my sight …
HE:
… Fury with the face of Heaven!
SHE:
… demon with the mask of love!
HE:
I curse …
SHE:
I shudder …
HE:
… the day that I set eyes on you.
SHE:
… at the thought that I ever adored you.
HE:
How can you dare …
SHE:
Have you the insolence …
HE:
… to look at me again?
SHE:
… to remain in my presence?

This continues for fifty-three pairs of lines, during which the couple have second thoughts about each other and reach a reconciliation, ending as follows:

HE:
Peace, dear eyes!
SHE:
Peace, loving mouth!
HE:
No more wars, apple of my eye!
SHE:
No more scorning, o sweet glance!
HE:
If thee alone I love …
SHE:
If this my soul adores you …
HE:
… Cupid comes back to life.
SHE:
… this be the end of all our discords.

(trans. Lovett F. Edwards)²⁹

For any reader or listener, the rhythmic repetitions constitute the main point of the scene. They provide a kind of melody, and it is not for nothing that Pandolfi's anthology moves eventually in the direction of opera buffa. Standing back a little, one notices that neither lover is in any position to listen to what the other is saying, since their thoughts and emotions proceed independently, if in complete parallel. There could be a source of parodic fun here, even a psychological comment on the self-absorption of lovers. Here again, however, questions have to be asked in relation to ‘improvisation’, and we must suspect that the overwhelmingly literary origins and content of so much of the Lovers' dialogue must have meant more learning by rote, and less creative freedom on stage, than was the case with many of the other masks. Nevertheless, long passages of material were available for constant recycling, in just the same way as the more fluid jokes of a Zani or an Arlecchino. Verbal and rhetorical constructs, sometimes of considerable complexity, have taken on the status of theatrical currency, in a bank which also deals in songs, jokes and pratfalls.

The appeal of some of the more ornate and abstract constructions remains hard to evoke for the modern reader or scholar. Theatre history is sometimes the study of an alien culture, and we have to record what was there whether we can empathize or not. We should make the comparison, perhaps, with the films of the Marx Brothers in the 1930s and 1940s. In these comic ‘texts’, which seem in so many ways to parallel the spirit of commedia dell'arte, it was felt necessary—presumably by the public as well as by the makers of the films—to suspend the anarchic mayhem at intervals, and give way to ten minutes of sentimental exchanges and a song from Alan Jones (or equivalent) and his leading lady.³⁰ Fashions in sentiment change, and what seems vacuous to us must have appeared more substantial in the past. To revert to less distant comparisons, the conceit-laden exchanges to be found in Isabella Andreini are not so different in kind from those in Love's Labours Lost or other admired Shakespearean dialogues. The difference is that Shakespeare had a lighter touch, and pursued his conceits with reference to the design of a single scripted play, rather than storing them for modular use in a succession of different scenarios.

It has to be faced, however, that no modern audience can respond to the kind of material which commedia dell'arte lovers used with anything other than boredom or derision. This means that even the most meticulous modern revivals of the genre, such as the hilariously successful productions of TAG Teatro,³¹ have to mock the verbal and emotional pretensions of the lovers and abandon any attempt at sentimentality. It is hard to say how far this choice is inherent in the material itself, and thus ‘authentic’, and how far on the contrary it responds to an irreversible change of taste in the theatregoing public.

Our survey of the repertoire used by ‘verbal masks’ is still provisional. We can nevertheless feel some confidence in proposing that actors who played these roles had to memorize a lot of fairly large-scale speeches and exchanges; and that their ‘improvisation’ consisted mainly in choosing and adapting what they had in their memories (and in their commonplace books) to fit the story being played. The longer the units of material are, the more we can perceive how literary sources, or at least written sources, contributed in significant measure to this unscripted, supremely theatrical medium. The links between commedia dell'arte and commedia erudita have, on this basis too, to be recognized as strong.

SCENARIOS

Pieces of repertoire and pieces of inspired invention, speeches, dialogues, songs, routines and slapstick numbers, everything had to be fitted on the night of a performance into the framework of a chosen story. The virtuosity of each performer must be allowed to flower, but the fictional patterns of the chosen fabula also had to be maintained. The story itself was based on a repertoire of stock units from ‘regular’ scripted comedy: family rivalries, love affairs, disguises, misunderstandings, tricks and discoveries. It was almost always divided into three acts, rather than five. At the same time it would seek to incorporate extra inventive elements, usually resolving themselves into single memorable comic scenes. An example would be Il cavadente (‘The Dentist’) from the Flamminio Scala collection. Here a large part of the first act is orchestrated to prepare for a scene just before the first interval, in which Arlecchino masquerades as a travelling dentist and unnecessarily extracts most of Pantalone's teeth. The remainder of the show goes on to explore different intrigues and gags, but the ‘dentist scene’ still gives the title to the show. The story to be told, the succession of scenes and conspiracies which gives shape to the evening's entertainment, is referred to now in English as the scenario—an Italian word which was not always the one most frequently used at the time, equally common terms being canovaccio and soggetto.

In practice, the scenario was a hand-written document which belonged (for the time being) to a given company of actors, and which would be available as aide-mémoire during rehearsals and performance. (That there were rehearsals is beyond doubt—company contracts mention the imposition of fines for missing them; and there would need to be a team consultation before each performance, to ensure that everyone was clear about the relationship of his/her mask to each of the others, the order of the scenes, and which pieces of concerted repertoire needed to be remembered and inserted.) There was no reason why this document should ever go to a printer: that would make material available to rivals in other companies, and perhaps reveal too many trade secrets to an audience whom one wanted to dazzle and mystify. Accordingly, nearly all the collections of scenarios which have survived from the seventeenth and eighteenth centuries are still in manuscript.³² Just one set was published: that of Flamminio Scala, who acted as the lover ‘Flavio’ and was a director of various companies. He published a group of fifty scenarios entitled Il teatro della favole rappresentative (‘The Theatre of Stories for Performance’) in 1611.³³ This collection has inevitably attracted more attention than any other. It is the earliest one to have survived, it comes from a distinguished practitioner of the ‘classic’ period, and it is the only one available to the general reader. Nevertheless, comparison with the manuscript collections does not suggest enormous difference between Scala and the rest.

The whole point of a scenario is that it does not give away much detailed information as to what is going to happen in each scene. It outlines the story, says who is on stage at any given moment, but then leaves spaces for the material, new or old, of the individual performers, using phrases like ‘they do their scene … he does his lazzi … they have their love scene …’ and so forth. Important plot information which has to be conveyed to the audience is spelled out in detail, because the actors need to remember it too, but the entertainment for which the plot is a pretext can often hardly be glimpsed. Granted the particular emphases of our present study, which has tended to focus on the single theatrical moment rather than on larger-scale plot structure, scenarios are a subject over which we shall not be able to linger. What can be said at present on the analysis of the single scene has been said already, in the uniquely valuable monograph of Tim Fitzpatrick (1985), who establishes that improvisation scenes always seem to break into a binary pattern. Either only two characters are involved at a time, or larger groupings seem to be resolvable into two opposing parties, with at the most a third person acting as umpire in the contest. This observation coincides well with the kind of material which we have been able to choose for comment: there have rarely in our examples been more than two characters foregrounded at one moment.

A scenario was above all a practical document, a set of guidelines and instructions for performers. At the start there is a list of characters, but also a list of necessary properties (‘Two boxes of sweets; a dentist's disguise; blacksmith's tools; a good armchair’, in Il cavadente). The characters are always listed by households, whereas in ‘regular’ scripted comedy it was usually either in order of appearance or in order of importance. The following is the cast-list for Il cavadente:

Pantalone
Orazio, his son
Flamminia, his daughter
Pedrolino, his servant.
Flavio
Isabella, his widowed sister
Franceschina, their serving maid
Arlecchino, their servant
Doctor, alone
Capitano Spavento, alone
Pasquella, an old woman on her own

This layout is important for the actors, when they study it. In last night's performance, Pedrolino may have served the Doctor, and Isabella may have been his daughter; in a different story, the Capitano might have had a house on stage, rather than being ‘alone’ and always entering from the street. Here they can remind themselves at a glance of tonight's status, relationships and stage territory. The subsequent narrative is also usually disposed on the page in such a way that actors can spot immediately the next moment at which they will be needed. Continuing with Il cavadente, the first scenes go as follows:

Pantalone tells Pedrolino of the love he bears to the widow Isabella, and how he fears that his son Orazio is his rival, so he has resolved to send him to university. Pedrolino disapproves, taking Orazio's part. They quarrel and come to blows: Pantalone attacks Pedrolino and bites him on the arm, making it seem a good hard bite. Pantalone leaves, threatening him, and telling him to talk to Franceschina on his behalf. He exits. Pedrolino: that he will get his own back for the bite he got from Pantalone. To him

Franceschina going to look for Orazio on her mistress's orders. She sees Pedrolino, and hears from him why his arm hurts. In revenge, they agree to pretend that Pantalone's breath stinks. Franceschina goes indoors, Pedrolino stays; to him

Flavio tells Pedrolino of his love, bumping into his arm. Pedrolino yells; then they agree to pretend Pantalone's breath smells. Exit Flavio, Pedrolino stays; to him

Doctor says he is owed 25 scudi by Pantalone …

(Il teatro delle favole rappresentative, Day xii)

In fact there is a typographical error here, because Pedrolino should be highlighted for the first scene—following the conventions normally used in the same collection, it should read:

Pantalone tells

Pedrolino of the love he bears to the widow … etc.

Otherwise the text is careful in spelling out exits and entrances, and also punctilious in using the phrase in questo or in quello—translated as ‘to him’, ‘to her’, or ‘to them’—indicating to a newly entering character that he does not have to wait for an empty stage. By contrast, other details are unexplained, because the actors know what is intended: in the above opening scene there must be a well-rehearsed sequence of slapstick comedy whereby a quarrel between master and servant leads to a bite on the arm. The decision to pretend that Pantalone's breath stinks seems quite arbitrary, and has to be handled by the actors as they see fit. Its rationale is in fact that it prepares for a lengthy sequence at the end of the act, involving a large number of the cast in quick succession—a good example of a scenario which calls for thought and team discipline, and where the structure is not entirely fragmentary:

… Pantalone hears how Pedrolino has talked to Franceschina; then hears Pedrolino say ‘Hey, boss, your breath stinks something awful.’ Pantalone laughs it off; to them

Franceschina does the same, saying that if his breath didn't smell, then Isabella would love him, and goes indoors again. Pantalone surprised; to them

Doctor arrives; Pedrolino signals to him about the breath; the Doctor does the same routine, and exits. Pantalone: he wants to ask his daughter if it's true about the smell. He calls her.

Flamminia admits to Pantalone that his breath does smell awful, and goes in. They remain; to them

Orazio from his house, confirms the same, then goes back in. Pantalone decides the tooth which is causing the stench must be pulled out. He orders Pedrolino to bring him a dentist …

(ibid.)

And this leads to the gruesome climax of comic sadism with the disguised Arlecchino, which is the set-piece finale of Act i:

… Arlecchino gets his tools out, all blacksmith's tools, giving them all silly names. He makes Pantalone sit down, and with a pincers pulls out four good teeth. Pantalone, in his pain, grabs the dentist's beard, which is false and comes off in his hand. Arlecchino runs away, Pantalone throws the chair after him and, moaning at the pain in his jaw, goes indoors, and this ends the first act.

(ibid.)

The rapid succession of characters all feeding the same story to Pantalone is a good example of laughter built up by accumulation. It also shows, in its repetitive variations on a theme, a version of our ‘modular’ structure on a slightly larger scale. One can imagine Orazio and Flamminia listening in the wings, picking up ideas and phrases from Franceschina and Flavio, and deciding whether to follow on by simply echoing what has gone before, by introducing a variant, or by doing something which is a total contrast. Once again the analogy with jazz is an attractive one: each soloist takes his or her turn and improvises round the established ‘melody’ in the light of what has just gone before. Ideally, they should also make up an ensemble.

Notes

1. The 1567 ‘season’ is narrated by Cesare Molinari (1985, p. 74).

2. The text is reproduced in Taviani and Schino, 1982, pp. 185-6; and translated in Richards and Richards, 1990, pp. 44-6. No woman is included among the partners of this early company. The first company contract to include an actress dates from 1564: see Taviani and Schino, pp. 183-4.

3. Major informative accounts have been produced by Pierre Louis Duchartre (English trans. 1929, repr. 1966), Kathleen Lea (1934), Allardyce Nicoll (1963), and Cesare Molinari (1985, so far only available in Italian). The translated documents of Richards and Richards (1990) are now indispensable for English readers. Duchartre and Molinari have a good selection of essential visual material; Lea and Molinari make the best critical use of available written sources, with Molinari obviously more up to date. Other important critical studies in Italian are those of Taviani and Tessari, listed in the general biography.

4. This means setting aside a number of apparent commonplaces about the genre, which date from eighteenth-century accounts, and especially from the version given by Luigi Riccoboni. Lea and Molinari show awareness of this problem, and distinguish between earlier and later periods: Duchartre and Nicoll are sometimes less discriminating.

5. In a comedy, Il teatro comico, which is a play functioning as a manifesto.

6. For a typical catalogue of a company, divided up in this way, see Molinari, 1985, p. 76.

7. It is tempting to think that the ‘classic’ series of engravings of the Recueil Fossard, reproduced by Duchartre and by Molinari, might actually record the entry of a stage Harlequin figure into a commedia dell'arte ensemble. The engravings have captions in French, and are believed to relate to a visit to Paris by the Gelosi company in 1577.

8. The story of Arlecchino and Brighella originating in the upper and lower towns of Bergamo is an eighteenth-century legend of relatively late fabrication.

9. Some ‘masked masks’, such as the ‘Beltrame’ created by Niccolò Barbieri, were nevertheless so individual as to die along with their impersonators.

10. Molinari is less disposed than I am to use the word ‘mask’ in relation to the Innamorati—it is a difference of semantic choice rather than of perception.

11. For a closer look at arte influences in Molière, see Andrews, 1989.

12. There is a study of the genre, plus texts of five comedies, by Luciano Mariti (1978). Kathleen Lea, with her usual acumen, identified the body of material independently, and lists many of the texts in her book of 1934, but she was not aware of the ‘Ridicolosa’ label.

13. Reproduced in full in Mariti, 1978.

14. Molinari (1985, pp. 77-82) shows how status was defined by whether a company performed in the open air (less respectable) or in a hired room or hall (more respectable).

15. Pandolfi, 1957, Vol. i, pp. 174-7; taken from Biblioteca Nazionale, Florence, D.4.6.23, no. 10 (Rari incunaboli palatini, Striscia 959). My own translated version of most of the dialogue appears in Andrews, 1991b.

16. The analogy with jazz is proposed independently by Molinari (1985, p. 40). It could well be a fruitful notion for critics to pursue: jazz, like commedia dell'arte, is in practice subject to severe stylistic restrictions, within which its creative improvisations flourish.

17. There is room for an enquiry into the frequency of modular scenes, and even of ‘elastic gags’, in Plautus: there is no reason why they should not have appealed quite independently to the professional actors for whom he wrote in the 3rd-2nd century bc.

18. Andrews, 1989, with the whole scene translated in an appendix.

19. One uses the term ‘family of masks’ in the sense that every Capitano was substantially different from every other—they had different names, no facial masks, and varied in stage personality as much as individual Lovers.

20. Antonio Pardi, Stupende Forze e Bravure del Capitano Spezzacapo e Sputasaette (Heredi di Giovanni Rossi, Padua, 1606), reproduced in Pandolfi, 1957, Vol. i, pp. 343-8.

21. Francesco Andreini, Le bravure del Capitano Spavento (Somasco, Venice 1607).

22. Giulio Cesare Croce, Vanto ridicoloso di Trematerra (Bartolomeo Cochi, Bologna, 1619), reproduced in Pandolfi, 1957, Vol. i, pp. 353-9.

23. Anon, Vocabulario Gratianesco; Biblioteca Nazionale, Roma, cod. Sessoriana no. 587. Reproduced in Pandolfi, 1957, Vol. ii, p. 32.

24. Anon, Le cento e quindici Conclusioni in Ottava Rima del Plusquamperfetto Dottor Gratiano … (s.l. 1587); reproduced in Pandolfi, 1957, Vol. ii, pp. 11-19.

25. A manuscript entitled Dialoghi scenici di Domenico Bruni detto Fulvio, comico Confidente …, Biblioteca Burcardo, Rome, cod. 3-37-5-35. Bruni also published Fatiche comiche (Paris, 1623) and Prologhi (Turin, 1621).

26. Isabella Andreini: Fragmenti di alcune scritture … (Tarino, Turin, 1621).

27. Quoted in Pandolfi, 1957, Vol. ii, pp. 58-9.

28. Translated from the Diario of Giuseppe Pavoni (Rossi, Bologna, 1589).

29. Oreglia 1968, pp. 119-22.

30. This refers not only to A Night at the Opera, but also to A Day at the Races, At the Circus, and The Big Store.

31. At the time of writing, this company under the direction of Carlo Boso has probably got closer than any other to reproducing the commedia dell'arte experience on a regular basis, with the full approval and enjoyment of modern audiences. It is based in Mestre near Venice, but has toured in other European countries.

32. The most important collections are listed by Lea, 1934 (Vol. ii, pp. 506-9) and Molinari, 1985 (pp. 42-3). Lea has a list of known scenario titles on pp. 509-54; perhaps not exhaustive, but very useful.

33. A modern edition is available: Scala/Marotti, 1976. Regrettably, the English translation by Salerno (Scenarios of the commedia dell'arte …) is so full of material errors as to be practically useless.

General Bibliography

Andrews, Richard. 1989 ‘Arte dialogue structures in the comedies of Molière’, in Cairns (ed.), 1989, pp. 141-76

Barbieri/Taviani. 1971 Nicolò Barbieri: La Supplica … (1634), ed. F. Taviani (Il Polifilo, Milan)

Duchartre, Pierre Louis. 1929 The Italian Comedy, trans. R. T. Weaver (Harrap; reprinted Dover, New York, 1966)

Lea, Kathleen M. 1934 Italian Popular Comedy, 2 vols. (Clarendon, Oxford)

Mariti, Luciano. 1978 Commedia Ridicolosa … Storia e Testi (Bulzoni, Roma)

Molinari, Cesare. 1985 La commedia dell'arte (Mondadori, Milan)

Nicoll, Allardyce. 1963 The World of Harlequin (Cambridge University Press)

Oreglia, Giacomo. 1968 The Commedia Dell'arte, trans. F. Lovett Edwards (Methuen, London)

Pandolfi, Vito. 1957 La commedia dell'arte, storia e testo, 6 vols. (Sansoni, Florence) [Anthology]

Richards, Kenneth and Richards, Laura. 1990 The Commedia Dell'Arte: A Documentary History (Blackwell, Oxford)

Scala/Marotti. 1976 Flaminio Scala: Il teatro delle favole rappresentative (1611), ed. Ferruccio Marotti (Il Polifilo, Milan) [Anthology of scenarios]

Taviani, Ferdinando. 1970 La commedia dell'arte e la società barocca. La fascinazione del teatro (Bulzoni, Rome)

Tessari, Roberto. 1969 La commedia dell'arte nel Seicento. Industria e arte giocosa nella società barocca (Olschki, Florence)