The purpose of this paper is to explore the influence of the invention and spread of printing on the reception of certain types of drama in France in the 15^th and 16^th centuries. In particular, I wish to examine to what extent the evolution observed in the writing and performance of religious plays during this period reflected a change from what has been called the manuscript culture of the Middle Ages to the print culture of the Renaissance and after. In particular, I shall try to establish whether the arrival of printing can be held to account for the transformation of the nature and status of drama. For, in the Middle Ages, drama was, in a very real sense, not literature; it was a visual and aural spectacle, which could only be "consumed" in a theatre. In the post-medieval period, drama has become one of several recognised literary genres. By this I mean that the literate public is able to read plays privately in their own homes, without going to the theatre; indeed, arguably nowadays drama is more read than seen in performance.

Mystery plays flourished in France especially between mid-15^th and the mid-16^th centuries. The chronological boundaries of this genre are not unconnected with contemporary political and social history. Such spectacles depended on a peaceful and reasonably prosperous economic climate; and this did not exist until the 100 Years' War was well and truly over. On the other hand, the decline of the genre is attributable to many factors, but religious censorhip amd changing tastes in the mid-16^th century were certainly two of the most influential.

The century during which mystery plays dominated the French stage coincides almost exactly with the discovery and spread of printing throughout Europe. The first printers came to Paris from Germany in 1470; by the end of the 15^th century there were printers and book-sellers established in most of the major French cities. By the early 16^th, Paris and Lyon were major European printing and publishing centres. What I want to look at is the relationship between these two phenomena, between the printing industry and the theatre in late medieval and Renaissance France.

Of the approximately 180 surviving mystery play texts, the majority are preserved in manuscripts. However, 36 have come down to us in printed form, although most these printed texts went through a number of different editions, resulting in a total of about 125 different editions of mystery plays. One play, Jean Michel's Mystère de la Passion, even went through over 20 different editions, spread out over 50 years. The earliest printed mystery plays dates from about 1484; the majority were printed between 1510 and 1542. However, perhaps significantly, some mystery play editions were published as late as 1630, well after mystery plays were no longer performed.

One surprising fact about the corpus of mystery play texts is that there is practically no overlap between printed and manscript plays; that is to say, virtually no mystery play survives both as a manuscript and as a printed edition. Is this just a coincidence? Before making any explicit comparisons, I shall start by talking about mystery play manuscripts and then say something about printed mystery plays.

French mystery plays manuscripts⁽¹⁾

are, in several major respects, very different from the manuscripts that preserve the texts of other medieval French literary genres. Most of the major chansons de geste, romances, lyric poetry and didactic works survive in numerous manuscripts. Philologists can usually reconstruct the often complicated relationships between the different manuscripts of these works, and produce complex stemmas to illustrate the manuscript tradition. In contrast to this situation, mystery play manuscripts are in almost every case unique. That is, only one manuscript copy of each mystery play survives, although two copies of the main manuscript are known to have been made for performances at Mons and Châteaudun. There are two notable exceptions to this generalisation, Arrnoul Gréban's Passsion and Jacques Milet's Destruction de Troye, where a half-a-dozen or more manuscripts survive, but these are special cases and can be accounted for.

Even so, play manuscripts were extremely varied in their presentation. Their diversity is at least as great as those of the other genres, if not more so. They can be written on large folios or very small narrow ones; they can be of parchment or of paper. They can be richly illuminated, with scores of miniatures, or semi-legible scrawls; they can be copied on one side only of the sheet of paper, or on both sides. The pages can be ruled and text evenly spread out, or the size of writing can vary, as can the number of lines per page. The writing can be in two columns per page or just one. There can be large numbers of stage directions (either in French or in Latin, either in the margin or in the centre of the page) or there can be none. The manuscript can be roughly contemporary with the play's composition, or much later. A manuscript can contain one play, or two, or six, or twelve, or forty, or seventy-three.

Why are play manuscripts different from the manuscripts of other genres? The answer, I would suggest, lies in the very specific function of a play manuscript. A medieval play was only fully realised in a performance; without a performance, the dramatist could not reach his audience. The mere writing out of a text did not permit communication between a playwright and the members of his public. With the other genres, however, the writing out of a text, especially in several copies, was all that an author needed or could hope for, towards the end of the Middle Ages, in order to ensure that his work might be read; for reading by one individual, either for his own benefit or for the benefit of several others, was the normal method of transmitting a text other than a play. The recently-published book by Joyce Coleman on public reading in medieval France and England demonstates this convincingly⁽²⁾

. Play manuscripts, however, were not written in order to be read; they were brought into existence in order to permit a performance of the play. Since virtually all mystery play and miracle play performances were unique, i.e. since it was very rare for severalÑor even twoÑperformances of exactly the same play to be organised, it is not surprising that play manuscripts tend to be unique as well.

All medieval play manuscripts were in some ways related to a specific performance; but not all such manuscripts served exactly the same purpose. The great variety of play manuscripts is explained by the wide range of functions that manuscripts might have in relation to a given performance. The best way to illustrate this is to imagine how a play is put on nowadays and to see how medieval practice must have been different. NowadaysÑand this has presumably been the case for some timeÑall the participants in the production of a playÑplaywright, producer, designer, actors, prompter, etc.Ñsimply purchase (or are given) a copy of the printed book, or a photocopy of a typescript; thus every participant has the complete text, i.e. the words for every actor, the stage directions, etc. Clearly, in the Middle Ages, it was not possible for every participant to have a complete copy (his own manuscript) of the play. The cost and time needed to produce so many copies would prevent this.

Contemporary evidence, including archive information as well as the manuscripts themselves, suggests that, as far as written texts are concerned, the preparation of the performance of a medieval play went through several stages.

(i) The dramatist starts composing his play, using several rough drafts.

(ii) The dramatist arranges for a fair copy of his final text to be written up; he either does this himself or gets one or several scribes to do the job. This fair copy, usually called the livre original or the registre, forms the basis for several other possible manuscripts.

(iii) A scribe copies out the actors' rôles, referred to variously as the roole or rollet or roullet; each actor is given a long, narrow manuscript which contains only that actor's lines. Each of his speeches is preceded by the last line spoken by another actor immediately before the speech; these are the cuelines. The actor uses his rle during rehearsalÑand possibly even during the performance.

(iv) A special, abbreviated copy might be written up for the producer, the meneur du jeu. This would expand greatly the stage directions, but reduce all speeches to two lines, the first and last of each rplique.

(v) After the performance, a final copy of the text is made, based on the author's fair copy; the purpose of this might simply be to keep a record of the event, or else to present the text as a gift to a patron or person of influence. Usually this type of manuscript is attractive, even luxurious, and often illustrated; it contains the text of the play, often laid out in double columns, but very few stage directions. It was intended primarily for consultation or private reading, and not meant to be used by actors or producers.

Of course, often in the late Middle Ages, a performance was not based on an original, newly-created text but on a previously existing one; in other words, somebody else's fair copy was borrowed, either from a neighbouring town or from a much earlier performance. In this case, however, the existing text was usually much revised, and a manuscript with numerous modifications written into it was the basis of the performance. Later, this now rather messy manuscript might be copied up into a new fair copy.

These various stages, not all of which were necessarily gone through for every performance of a mystery play, help to explain both the uniqueness and the great variability of play manuscripts.

Not surprisingly, printed editions of mystery plays were very different from these manuscripts. On the other hand, they were in most important respects very similar to other sorts of printed editions published in the first hundred years of printing. For general information on the early history of the printed book in France, I would refer the reader to the standard works on this subject⁽³⁾

. What particularly interests me in this paper is to see how the printed book influenced and/or changed mystery plays performances. The kind of questions that one might ask are the following:

1. How did printers obtain the play manuscripts that they published?

2. Is it possible to determine the exact purpose of printed mystery plays (e.g. reading or performance)?

3. To what extent did printed play texts circulate around France?

4. When and where were printed play texts performed?

5. How were printed plays performed; in particular were they re-converted into manuscripts in preparation for performances?

6. How many copies of each play were printed?

7. How much did they cost? Could an ordinary person - like an actor in a mystery play - afford to buy a copy for himself?

Book-sellers obviously existed before the invention of printing; during and after the spread of printing, however, the nature and extent of their business must have changed significantly. For quite a long period, these merchants sold both manuscripts and printed books. One of the best examples I know of is the so-called "libraire de Tours"⁽⁴⁾

. A catalogue of the books that this book-seller from Tours was trying to sell has survived; it dates probably from the very end of the 15^th century or just after. Most of 265 books in the catalogue were in manuscript form; 28 were printed books, "livres en impression". Mystery plays account for 34 of the manuscript books. A significant number of the titles of the manuscripts in this catalogue subsequently appeared in printed editions bearing the name of the famous Parisian publisher Antoine Verard. It seems probable that the Tours book-seller was one of Verard's suppliers of manuscripts. This would suggest that book-sellers bought up manuscripts and sold them on to the printing industry. Of course, publishers themselves also used their own initiative in getting hold of potentially good-selling manuscripts, which they then published. As evidence, one can point to the fact that many printed editions of mystery plays state on their title page that the play has recently been performed to great acclaim. Clearly, a successful performance could lead very rapidly to a published edition of the text, as indeed is sometimes the case today.

To produce the first printed edition of a pre-existing text, a printer needed to acquire a copy of the manuscript. This manuscript was then heavily marked, annotated and scribbled over, in order to give the compositor instructions about character-choice, page-divisions, running-titles, etc. This preparatory process virtually destroyed the manuscript itself; after the process of typographic composition it would have been quite illegible. Now, as I suggested earlier, virtually all mystery play manuscripts were unique. It is therefore not surprising that none of the play manuscripts that were converted into printed editions have survived. One can see, therefore, that the uniqueness of mystery plays manuscripts and the nature of the printing process account for the lack of overlap between manuscripts and printed editions of mystery plays.

I will now attempt to tackle those questions listed above relating to the reception of the editions, i.e. print-runs and prices, to their circulation throughout France, and to their function. One possible change that one might expect to find in the organisation of play performances, as a result of the increasing presence of numerous printed editions of mystery plays, is that all the major participants in a play would be expected to buy, or to be given, their own copy of the text. However, for this to happen, there would need to have been plenty of copies available; some of the large-scale plays required over 200 actors. Moreover, the price of the editions would have to be low enough for the average citizen to buy one; most actors in mystery plays were not professionals, but amateurs, ordinary middle-class or lower middle-class citizens.

To explore this problem, we need information about print-runs and prices. Unfortunately, these are not subjects about which specialists in the early history of the printed book can provide much precise information. As regards early 16^th-century Paris, one of the main authorities on the question, Jeanne Veyrin-Forrer⁽⁵⁾

, hypothesises that the print-runs of books of the sort we are interested in were usually of the order of either 600 or 1200 copies. These figures are based on the number of copies of a composed folio that could be printed in a day, depending on the number of presses available. Most printers rarely had enough characters of each type-set to print more than one or two folios at a time. After the required number of copies of each folio were printed, the printing form was broken up, the typographical characters were put back into their cassetins, and only then could the composition of the next folio begin. I believe that there are good grounds for accepting these figures, although time will not allow me to provide the evidence here. They mean, in practice, that about 600 copies of most editions of mystery plays were printed; in some exceptional cases, up to 1200 may have been printed. These print-runs seems quite large, at least as large as those of present-day academic books. I wonder, will the publishers print 600 copies of the acts of this conference?

However, we must not forget that many mystery plays were printed several times. These re-editions were in theory simple reprintings of earlier editions. But for reasons I have already explained, each new edition had to be recomposed folio by folio; so, owing to human fallibility, second impressions were never absolutely identical to first impressions, even if they set out to be. The Mystère de la Passion by Jean Michel was reprinted over 20 times between 1486 and 1542. This was clearly one the best-sellers of the period. Even if only 600 copies of each impression were produced, that would mean that 12,000 copies of the play were circulating in France by the mid-16^th century. There seems no doubt that there was a sufficient number of copies available to satisfy any possible demand. In any case, if more were needed, plays were rapidly reprinted.

Nevetheless, even if there were plenty of copies available, were they cheap enough for the average citizen to buy? This is an even more tricky problem than print-runs. In the 16^th century books did not have fixed prices. They were usually sold unbound; so the cost of any binding was an additional variable. Most information about early 16^th century book-prices is anecdotal rather than systematic, and needs to be treated with caution. One apparently systematic source of information is the inventories of books which were sometimes drawn up when a wealthy book-seller died. Several of these survive. In these inventories, professional valuers draw up a list consisting of all the different works found in the printer's shop, and giving their titles, their length in folios, the number of copies unsold of each work, and the valuation in the local currency. Such a list is found in the inventory⁽⁶⁾

made after the death of Jean Janot, a famous Parisian printer and book-seller, who published a number of mystery plays between 1511 and 1519. His books were valued at a uniform rate, which was based on the number of printed folios in each book. The rate was 10 sols parisis for 500 printed folios; a folio was normally folded two or three times to produce respectively gatherings of four leaves in-quarto or eight leaves in-octavo. This meant that, for example, each copy of a 3,500-line mystery play called the Mystère de l'Assomption, which consisted of 20 leaves in-octavo, was valued at only 2½ deniers parisis, i.e. 2½ pence. However, it is certain that the prices attributed to Janot's books by the valuers were far below their retail market value - perhaps only a quarter. This, by the way, was also the case with almost everything else in the inventory, like clothes, furniture, jewellery, printing equipment, etc., which, at best, were assessed at only half or less of their true value.

Another, more reliable, source of information is that provided by Hernando Colón (Fernando Columbus, Fernand Colomb). Hernando, the son of Christopher Columbus, was, among many other things, an avid book-buyer. He bought a considerable number of French books in Lyon in the 1530s, including several mystery plays. Every time he bought a book, he wrote a little note on the inside cover, stating where and when he bought the book, how much it cost, and the rate of exchange between the local currency and the Spanish ducat. For a copy of the same mystery which was valued at 2½ deniers in Jean Janot's inventory, the Mystère de l'Assomption, Colón paid 20 deniers, but his copy was already bound. From this and other similar examples, one can conclude that it would probably have cost between 10 and 20 deniers to buy a small to medium-sized book published in Paris in the 1520s and 1530s. At this time, a master carpenter might earn two or three sols per day (i.e. 24 or 36 deniers), and a labourer about two sols per day. These figures are obviously approximate, but they do suggest that, for most people, a day's work would earn more than the price of a book.

On the basis of the above calculations, it seems certain that most, if not all, participants in mystery plays would, if the text had been printed, have found a plentiful supply of copies, and that these would have been available at affordable prices. However, this is very far from saying that the participants did buy them. In fact, it seems that they did not.

The evidence for this conclusion is varied, and it also casts light on the question of the circulation of these editions. If we take the best-selling mystery play, Jean Michel's Passion, it is very clear that it circulated widely throughout the French-speaking territories. However, no manuscript of this play has survived; there are only printed editions. It was originally performed in Angers in 1486, and published immediately afterwards. Subsequently, it was performed at Amiens in 1500 and in Mons (now Belgium) in 1501. It was also used at Valenciennes in 1547, and several times in the Savoie and the Dauphiné at various dates in the 16^th century. However, our information about these performances, and about the texts used at them, comes not from printed editions of the Jean Michel Passion, but from manuscripts. In every one of these cases, the text used was a manuscript, which was a hand-written copy of the printed edition. Of course, these manuscript versions freely adapted the original, cutting and adding scenes, characters, staging effects, etc. Although there is no doubt that the source of these plays was one of the printed editions of Jean Michel's Passion, the version actually used at the performance was not the printed edition itself.

Furthermore, there are many examples of late French mystery plays, which have survived in manuscript form, but which incorporate large sections of earlier printed editions. One of the best examples is the Parisian Mystère de Saint Denis, the manuscript of which dates from about 1530⁽⁷⁾

. It includes complete, and virtually word-for-word, the play I have already mentioned twice, the Mystere de l'Assomption published in Paris by Jean Janot between 1511 and 1519, and bought by Hernando Colón in Lyon in 1535.

Moreover, if one were to compare the printed editions of mystery plays with the various types of manuscript described earlier, the one they resemble most is the final copy, the one that was carefully written, often laid out in double columns and illustrated, and containing few stage directions - one that was designed for reading and not for an actor or a producer. This type of manuscript was the only one that followed, rather than preceded, the performance of the play in question. Similarly, as I have already noted, many printed editions of mystery plays were published after, rather than before, the performance of the play. Indeed, they often mentioned the recent performance on the title page, no doubt as a way to boost sales⁽⁸⁾

.

The conclusion to be drawn from the above evidence seems to be that the performance of mystery plays did not change in any fundamental way as a result of the spread of printing. The latter, however, did have important consequences as far as the circulation and survival of the plays were concerned. Most mystery plays were printed in Paris; a small number were also published in Lyon. Of these, the majority were sold and remained in these major printing centres. But some of the most frequently printed mystery plays, like the Gréban-Michel Passion, the Mystère du Viel Testament and Mystère des Actes des Apôtres, spread far and wide, and are known to have been the inspiration for performances in the remotest corners of the French-speaking world - for example, at the far end of valleys high in the Alps - and at dates as late as the 17^th century in places like the Auvergne.

For although, in the main cities of France, performances of large-scale mystery plays were gradually suppressed and even those of small-scale plays grew progressively rarer, printed editions of the plays continued to be published throughout the 16^th century and the first half of the 17^th. But, except in the more marginal regions, these were texts designed to be read, and probably formed part of the devotional reading of the non-intellectual, but pious public. As spectacles, the mystery plays had ended, though belatedly, perhaps, and well after what most people consider to be the end of the Middle Ages. But, thanks to printing, many mystery play texts continued to circulate during the 16^th and even into the 17^th centuries, reminders of a theatre very different from what we now consider to be French Renaissance and Classical drama. Indeed, it seems to me a highly significant fact that the two of the very last printed editions of French mystery plays were published by the Nicolas Oudin, who initiated the famous collection of popular littérature de colportage, the Bibliothèque Bleue.

Graham A. Runnalls (University of Edinburgh)

^1.G. A. Runnalls, "Towards a Typology of Medieval Play Manuscripts", The Editor and the Text : Essays in Honour of A.J.Holden, Edinburgh University Press, 1990, 96-113.

2.

^2. Joyce Coleman, Public Reading and the Reading Public in Late Medieval England and France, Cambridge University Press, 1966.

3.

^3.For example, H. J. Martin (éd.), L'Histoire de l'Edition Française : Le Livre Conquérant, Paris, 1982; A. Parent, Les Métiers du Livre à Paris au XVIe siècle, Genève, 1971; G. A. Runnalls, "The Book market in early sixteenth-century France", Early Book Society Newsletter, Second series, vol. I (1996) 3-5.

4.

^4.G.A. Runnalls, "The Catalogue of the Tours bookseller and late Medieval French drama", Le Moyen Français 11 (1982) 112-128; and "The Catalogue of the Tours Bookseller and Antoine Vérard", Pluteus 2 (1984) 163-174.

5.

^5. J. Veyrin-Forrer, "Fabriquer un livre au XVIe siècle", dans: Histoire de l'Edition Française : Le Livre Conquérant, Paris, 1982, pp. 279-301.

6.

^6.Paris, Archives Nationales de France, Minutier Central CXXII, 4 (Etude de Jehan de Calais, Petit-Pont) [ANF Cote: MC: ET/CXXII/4]. I am in the process of preparing a critical edition of this document, in collaboration with Stéphanie Rambaud.

7.

^7.G.A. Runnalls, "Un siècle dans la vie d'un Mystère: Le Mystère de Saint Denis", Le Moyen Age XCVII (1991) 407-430.

8.

^8.However, there is good reason to believe that some of these alleged recent successful performances were fictional.