‘Look! […] Things People Can’t See!’ Wordbooks, Reader-Listenership, and Invisible Theatre in Handel’s Oratorios

Abstract

In eighteenth-century England, anyone attending an opera, an oratorio, or even a church service would typically have had a printed ‘wordbook’ made available to them to read during the performance. Such wordbooks, whether available for purchase or distributed free of charge, contained the words to be sung (the libretto), usually with translations if necessary, and sometimes also explanatory footnotes, prefaces or plot summaries, or lists of dramatis personae. Examining several oratorios of George Frideric Handel, especially Saul and Theodora, this article asks how the wordbook influenced the drama of a performed work and to what extent this impact made it necessary to have an actively reading audience. The article also explores the use of stage directions in oratorio wordbooks, arguing that they provide rich opportunities for the audience’s imagination by suggesting images that the performance alone cannot provide (since English oratorio probably included no stage action). It notes the wordbook’s necessity in determining which singer is portraying which character, as well as the expressive and dramatic use to which these character identifications can be put. And it compares the practices of different oratorio librettists, suggesting great sensitivity to the unique imaginative power of the oratorio-with-wordbook medium.

1. Introduction

On the surface, Handelian oratorio might seem like an artform rooted in compromise; watered-down opera lacking the spectacle that made its sister art so popular. A contemporary of Handel certainly thought so: attending an oratorio performance, that anonymous contemporary ‘found this Sacred Drama a mere Consort, no Scenary, Dress or Action, so necessary to a Drama’ (Burrows et al. 2015, p. 535, quoting Anonymous 1732).1 Even making allowances for the exaggeration common in satire, that anonymous spectator might seem to have had a point. Why go to the theatre for a ‘mere Consort’, with none of the frills or thrills of staging?

There are many possible reasons. Being (usually) sacred music-dramas, oratorios circumvented the ban on opera performance during Lent, and were thus as close as the public could get to opera during the Lenten season. Neither should we underestimate the appeal of English-language music-drama (even unstaged) to the English, nor the sincere desire to experience sacred music-drama for spiritual reasons (even though the ban on publicly staged biblical drama forced the oratorio to eschew ‘Scenary, Dress’, and ‘Action’). But not all oratorios were intended for Lenten performance, not all oratorios used sacred subjects, and the English did attempt opera in their own language opera during the eighteenth century. Perhaps, then, we should rephrase the initial question: since we know that many people did indeed engage with oratorio, what can we uncover about the uniqueness of that genre, that engagement, and the aesthetic experience of those people?

In the following article, I address that question, arguing that oratorio’s lack of staging comes into focus as an imaginative resource when we consider certain particularities of how such drama was consumed, and prepared for consumption. In order to understand this resource, I examine reader-listenership and the phenomenon of the wordbook, exploring how an audience’s ability to read printed libretti before, during, and after the performance could influence their perception of the performed work. I will make the case for an actively ‘reading audience’, and raise the important question of whether wordbooks therefore constitute a crucial part of Baroque oratorio performance. I will not assert that every audience member read actively, grasped the wordbook’s importance, or even enjoyed oratorio as anything more than a stopgap measure. Neither do I claim that reader-listenership was the exclusive province of oratorio audiences (we shall see that wordbooks were also a feature of opera consumption). But I hope to show that the eighteenth-century English culture of reader-listenership offered particularly rich and fertile ground for imaginative engagement, when brought to bear upon an unstaged dramatic medium such as oratorio. And many oratorio creators seem to have been aware of this dramaturgical resource.

2. Results and Discussion

2.1. Reader-Listenership and Precedent for Oratorio as Unstaged Drama

I define ‘reader-listenership’ as the consumption of vocal music by both listening to it and reading its text, often at the same time (Twomey 2020, vol. 2, p. 85). And in eighteenth-century England, attendees of opera, oratorio, or even a church service would have a printed ‘wordbook’ available to them, containing the words sung (the libretto), often with translations if necessary, and sometimes with explanatory footnotes, prefaces or plot summaries, or lists of dramatis personae. Wordbooks had been a staple of English music drama since at least 1653, when the masque Cupid and Death was performed in London before the Portuguese ambassador, and the masque’s text (by Shirley) was published in conjunction with the performance (Shirley 1653).2 And booksellers also published wordbooks for productions of Handel’s Italian-language operas (as well as for his oratorios), selling them at theatres so that audiences could read them during the performance. The ubiquity of this practice is evidenced by casual and incidental references in journalism and even fiction; Aaron Hill mentions that, when attending an Italian opera, ‘I had my Book, and my little Wax-Candle’, the latter providing light by which to read the former (Hill and Popple 1734), and Henry Fielding’s novel Amelia also features an episode in which the titular protagonist attends a Handel oratorio, where a gentleman buys her a wordbook and candle, holding the candle for her during the performance (Fielding 1752, vol. 2, p. 81).3

At first glance, the concept of a wordbook seems unremarkable. Concert programs are common in the twenty-first century; opera audiences often have the opportunity to purchase them for reading during the performance; and sometimes they contain the libretto for the concert’s vocal music. But a modern concert program usually clarifies a piece’s history, articulates its creators’ aims, or aids with audience comprehension. Eighteenth-century wordbooks certainly did many or all of these things, but reading the wordbook did not enrich the listening experience in these ways alone; rather, it seems to have been an essential part of that experience. Wordbooks feature numerous dialogic relationships with the music dramas for which they were printed. And of these, perhaps the most important is the wordbook’s ability to suggest images, events, or even blocking (positioning of performers relative to each other on the stage) that the performance alone cannot present. Oratorio wordbooks often include detailed stage directions, which, combined with programmatic music and the characters’ words (both heard and read), appear designed to evoke vivid stage pictures in the imagination of the audience for what was always intended to be an unstaged drama.4

Again, this is not unique to oratorio, for spoken theatre also made imaginative use out of an apparent lack of visuality. Drawing on States (1987, p. 56), Wagner (2013) notes the ‘productive “tension between seeing and hearing”’ in Elizabethan drama, which did not use scenery in its stagings, but featured detailed scenic descriptions in its spoken text. Vandiver (2000) has explicitly compared this to theatre in classical Greece. During the English Commonwealth, however, playwrights had begun exploiting that tension even more keenly. The Commonwealth government officially prohibited the staging of plays, but play publication continued virtually unchecked; and these ‘closet dramas’, designed for reading rather than staging, incorporated highly detailed stage directions. Indeed, writing about the dramas of Cavendish, Straznicky (1995, p. 367) notes the presence of ‘protracted stage directions’ that ‘range well beyond the usual basic indicators of action to encompass such matters as costume, scenery, mood, motivation, attitude, and even narrative details from earlier in the play’. Straznicky (1995, p. 367) further asserts that these ‘descriptive, as opposed to imperative, directions […] even resemble in several of the plays the introspective techniques of prose fiction.’ Oratorio blended these two approaches, but in an unprecedented way. As in both closet drama and Elizabethan theatre, there was no scenery. As in closet drama (but unlike in Elizabethan theatre), scenery (along with action and costuming) was conjured by stage directions read by (not enacted before) the audience. But as in Elizabethan theatre (and unlike in closet drama), there was a performing cast on a stage.

2.2. Stage Pictures, Character Movement, and Props in the Inner Eye of the Reader-Listener

The following direction is from Belshazzar (Jennens 1744, p. 19), the appearance of the famous ‘writing on the wall’.5

• As he is going to drink, a Hand appears writing upon the
• Wall over against him; he sees it, turns pale with Fear,
• drops the Bowl of Wine, falls back in his Seat, trembling
• from Head to Foot, and his Knees knocking against each other.6

Importantly, no character actually says that a hand has appeared and is writing on the wall. Thus, without the wordbook’s vivid directions, audience members would have no idea why terror has suddenly beset the characters.

This looks like an oversight, but actually seems to be a hallmark of Jennens’s (the librettist’s) oratorio dramaturgy. The technique is also observable in Jennens’s (1738) libretto for Saul, written several years earlier. At three points in the course of the story, the title character, King Saul, throws his javelin in an attempt to kill a member of his court (Jennens 1738, pp. 9, 15, 18). The second such incident take place ‘offstage’, recounted by the near-victim David (Jennens 1738, p. 15). The first and last, however, occur in performed scenes, and in neither case do the sung words mention the incident (Jennens 1738, pp. 9, 18). The wordbook stage direction is the only way that an audience could the fact of the attacks (although in the first case a programmatic flourish in the music does assist): the stage direction ‘throws his Javelin’ appears in the manuscript score, in brackets (Handel 1738, f. 54v).

Later, Saul sends the character Doeg to fetch David from his home. David’s wife Michal stalls for time, claiming that her husband is sick; when this does not work, Doeg demands that Michal ‘Shew me his [David’s] Chamber’ (Jennens 1738, p. 16). There follows immediately the stage direction ‘David’s Bed discover’d with an Image in it’, and Doeg reprimands Michal for defying Saul (Jennens 1738, p. 16). The direction is also unusual in itself, not depicting the action of a character, but an imagined stage effect of some sort; if the ‘Bed’ is to be ‘discover’d’, we are perhaps to envision the lifting of a flat (piece of scenery). Johnson’s dictionary defines ‘image’ as (among other things) a statue-like ‘representation’ of an individual (Johnson 1755, p. 1047). The stage direction implies that David fled his home, and that Michal placed a vaguely human-shaped bundle in David’s bed (to make Doeg believe that David was still in the house).7 But without this direction, this fact, and Doeg’s reason for becoming angry, is not clear. This example is particularly subtle, and shows the librettist truly acting as ‘director’ of his audience’s imagination; Anthony Hicks’s description of oratorio as ‘a living drama played out in the ideal theatre of the mind’ (Hicks 1997, p. 156) seems particularly apt in such moments. But visual subtlety of this kind is not the only interesting feature of Jennens’s wordbook dramaturgy.

Later in the same oratorio, Saul appears ‘disguis’d at Endor’ (Jennens 1738, p. 19). The directions’ mention of his ‘disguis[e]’ adds enormous tension to Saul’s subsequent encounter with the Witch of Endor: without that direction, an audience member might have assumed simply that the witch was unaware of what the king looks like (despite the fact that this king made witchcraft punishable by death).

Finally, when David orders an execution, the wordbook informs reader-listeners that David’s instructions to ‘Fall on him, smite him, let him die’ are directed ‘To one of his Attendants, who kills the Amalekite’ (the ‘him’ of the command) (Jennens 1738, p. 21). No reference is made to the Amalekite’s fate in the sung text, so, without the direction, the audience might think that David’s command was ignored, that the man escaped the court, or even that David was simply raving (talking to himself, rather than ‘To one of his Attendants’).

Not all librettists used their stage directions in the same way, and Theodora—words by Thomas Morell (Morell 1750)—contains a particularly interesting set of stage directions that play a much more subtle role in the dramaturgy. When Didymus arrives to rescue Theodora from her ‘Place of Confinement’, he starts the scene ‘at a distance, the Vizor of his Helmet clos’d’ (Morell 1750, p. 14). After soliloquizing briefly in recitative, Didymus describes Theodora’s semi-conscious daze; then follows an aria sung while ‘Approaching her’, in which he professes desire to shield her from harm (Morell 1750, p. 14). When the aria concludes, a frightened Theodora cries out (in recitative), ‘Starting’, as the directions tell us, only for Didymus to declare (also in recitative) that he has come to free her, not to harm her (Morell 1750, p. 14). Not until the end of this last speech does Didymus reveal his true identity, and he does so both in words (stating his name), and also in gestures unseen, but likely read by the audience: ‘Discovering himself’, a rather vague description but probably meaning that he lifts the vizor of the helmet or removes it completely (Morell 1750, p. 15).

In contrast to the Belshazzar example, this scene is coherent even without its stage directions.8 There is no essential need to state why Theodora did not recognize Didymus; her ‘place of confinement’ may simply be dark. He seems to have startled her out of a daze in a place where she is being held against her will, so her initial shock is understandable; without the directions, audiences might reasonably assume that Theodora recognized him as a friend once that shock had passed. But the stage directions add great power to the interaction. Movement around the stage would probably have been possible even in concert presentation, but the stage directions obviate the need. Morell carefully choreographs Didymus’ approach to Theodora, as well as the build-up to the revelation of Didymus’ identity, placing the revelation at the end rather than the beginning of Didymus’ speech to Theodora by specifying this as the moment when Didymus reveals his face. Tension is thus maintained far longer than it might have been without the vizor (which, we should remember, is mentioned only in the stage directions). Instead of the simple progression that would have occurred without the vizor (Didymus’ tentative approach, then Theodora’s momentary startlement, and her immediate reassurance and recognition), the scene creates two layers of tentativeness (the recitative sung ‘at a distance’, the aria as he is ‘[a]pproaching’ Theodora), then presents the moment of startlement (Theodora ‘Starting’ and being afraid), before moving on to reassurance (Didymus stating his friendly intent) and finally full recognition (‘Discovering himself’). See (Morell 1750, pp. 14–15).

At first glance, it seems curious that Morell confines mention of the vizor to the stage directions and description of Theodora’s daze to the dialogue. There is perhaps an element of decorum in this. The stage directions can deal with merely practical matters that might be difficult to fit into poetic language, while poetic diction can maintain vagueness on the nature of Theodora’s condition. Whether this objectifies the female protagonist further (by making her a thing to be described), adds mystique or even mysticism to her daze, or simply encourages the audience further to use their imagination remains a subject open for debate.

2.3. Non-Specification of Events and Characters

Absence of stage directions can become particularly significant in the presence of descriptive headings. With two exceptions, the wordbook for Saul (Jennens 1738) never mentions a specific place in which the action occurs; scenes are headed ‘SCENE II. JONATHAN and DAVID’ (Jennens 1738, p. 11), or ‘SCENE IV. DAVID &c. To them an AMALEKITE’ (Jennens 1738, p. 21).9 But in nearly every scene, the wordbook does clarify which of the characters sings which lines of text. This may be a conscious choice on the librettist’s part to make the context as abstract as possible, in order to establish and fully embrace the absence of staging, and also to focus attention on character rather than location (a sensible choice, since he had performers, but no scenery). In the oratorio’s opening scene, however, the words are not stated to be sung by any named character at all. Instead, the heading ‘An EPINICION, or Song of Triumph, for the Victory over Goliah [sic.] and the Philistines’, and the simple numbering of its stanzas from I to V, make the scene into a sort of abstracted prologue, an impersonal and generalized expression of triumphant sentiment almost like a church anthem (Jennens 1738, pp. 4–5). It is as if Saul, that consummate ‘SACRED DRAMA’, has adopted (if only for its first few minutes) the expressive strategies of the later scripture oratorios like Israel in Egypt (Jennens[?] 1739) or Messiah (Jennens [1992] 1743).10 These latter two texts were, respectively, definitely and likely compiled by Charles Jennens, the librettist of Saul; perhaps the strategy was taking shape in his mind as early as 1738.

This anthemic non-scene’s place at the beginning of the oratorio may also be significant from a dramaturgical point of view. Belshazzar ends with an explicitly titled ‘Anthem’ of thanksgiving, but by that point the dramatic nature of the work had already been cemented (Jennens 1744, p. 32). The audience could decide that the drama had dissolved, the actors, so to speak, having removed their masks to address the audience directly ‘as themselves’; but the shadows of the characters enacted, and of the drama portrayed, would still invest the anthem. Saul’s Epinicion sounds before any named character has sung in any named place, and the oratorio thus represents, uniquely in Handel’s English output, a combination of the dramatic and the non-dramatic, specific and general, within a single work. It also moves back to the non-dramatic and the general near its conclusion; the fifth scene of the third act is headed ‘ELEGY on the Death of SAUL and JONATHAN.’, again with no specifications of location, and no identification of character in the distribution of its merely numbered stanzas (Jennens 1738, pp. 22–23). We might compare the oratorios of J.S. Bach, where, after a significant moment in the drama, a chorale often comments upon the action and shows more deeply the emotions felt by the characters (as well as the audience, since these chorales often serve as a kind of audience surrogate) as the story unfolds.11

More subtly, the wordbook for Saul is interesting in its treatment of the chorus, generally keeping them off to one side of the drama. With one exception (‘CHORUS of Women’), their texts are headed merely ‘CHORUS’, not ‘Chorus of Israelites’ or ‘Chorus of Attendants’ as they might be in other wordbooks.12 Nor, as other wordbooks might, does Saul’s announce their presence at the beginning of scenes in which they sing. For instance, the fifth scene of the second act gives its characters only as ‘DAVID and MICHAL’, yet, after both of these characters exit, the scene concludes with choral praise of David (Jennens 1738, pp. 14–15). Thus, even more so than in Greek tragedy, the chorus is remote from the story, commenting upon rather than participating in it.13 They even begin the second act with a scene entirely to themselves, devoid of any stage directions or descriptions, moralizing upon the dangers of jealousy: ‘ENVY! eldest-born of Hell!’ (Jennens 1738, p. 11).

Saul’s hovering between drama and abstraction is most obvious in its Epinicion and Elegy, but almost every chorus leans away from the concrete and even the situated. This is not a condemnation; it adds far greater texture to the weave of the dramaturgy. Other oratorios take pains to integrate the chorus into the scenes in the stage directions; Susanna (Mendes[?] 1749a) represents this approach very well. But Saul takes full advantage of the unstaged dramatic medium, almost zooming in and out, with the soloists for the most part enacting the story and the chorus mostly judging it like omniscient commentators (not narrators; they convey no new information). Dramaturgical fluidity is increased by relatively unpredictable distribution of labor in this respect. Sometimes the chorus does embody characters in the story (‘CHORUS of Women’). And there are solos in the Elegy and the Epinicion; when the alto soloist who has sung the role of David calls Jonathan ‘my Brother’ and cries that ‘Great was the Pleasure I enjoy’d in thee!’, we are almost certainly to think that the soloist has reverted to portraying David.14 Yet in this moment of greatest abstraction, and without a character heading for the passage, are we sure?

2.4. Librettist-Specific Dramaturgy-by-Wordbook

Far from being ‘a mere consort’, oratorio capitalized on its lack of stage action to suggest images just as spectacular as those brought before the eye by the costumes, sets, and machines of opera. Hicks asserts that vivid evocation of stage action would raise and disappoint audience expectations, and in certain instances this might very well have been the case (Hicks 1997, p. 162). But this seems to ignore the precedent for such stage directions in closet drama, and fails to acknowledge the pervasiveness of such directions. If they were a mere frustration, why continue to employ them? Indeed, it may be that the freedom to conjure images entirely through music and stage direction emboldened the librettist; like a novelist as opposed to a screenwriter, the oratorio (as opposed to opera) librettist did not need to worry whether the extravagant visual effects woven into the story were actually practicable. For all the ingenuity of Baroque stagecraft, it is difficult to imagine how machinists could convincingly have created the illusion of a disembodied hand accurately writing Hebrew script on the scenery. And it would have been difficult for Didymus to sing an aria through the vizor of a helmet.

Some librettists undoubtedly made greater use of these directions than others. Jennens seems to have relied heavily upon them to convey details of the plot, and sometimes to set the scene with remarkable specificity. The description of ‘The Camp of Cyrus before Babylon. A View of the City, with the River Euphrates running through it’ does more than establish the location; it conjures up a scenic backdrop that would almost certainly not have been present in the theatre during the performance of the oratorio (Jennens 1738, p. 6). Thus, the description makes this a meaningful background for the dialogue that unfolds in this scene (Act I Scene II of Belshazzar). Stage directions are a more peripheral tool in Morell’s oratorio dramaturgy, used to heighten the impact of self-evident actions rather than for the autonomous narrative propulsion seen in Jennens. Thomas Broughton is arguably subtler in his libretto for Hercules (Broughton 1745). One exchange between Lichas and Dejanira takes place as the latter is ‘Going’ out of the room, adding a strong aura of impatience to her lines and a greater desperation to Lichas’ aria, in which he tries to dissuade her from believing that her husband has been unfaithful (Broughton 1745, p. 21).15 Stage directions also make clear that Iöle (with whom Dejanira had been arguing in the previous scene) remains present throughout Dejanira’s subsequent recitative exchange with Lichas, the latter’s subsequent aria, Dejanira’s parting shot in recitative, Lichas’ recitative response, and the chorus that follows (Broughton 1745, pp. 21–22). Like a mute character on the Ancient Greek tragic stage, Iöle is relegated to, but steadfastly held in, the scene’s background, and it is the wordbook that accomplishes this effect with its stage directions.16

In other cases, invisible staging is employed to underline the drama’s larger themes. The title character in Samson (Hamilton 1733), for instance, is blind throughout that oratorio, so the audience cannot see the events that unfold around him any more than he can. Although this thematic relevance of ‘invisible staging’ might seem like something of a stretch, we should recall that the work on which this oratorio was based, Samson Agonistes, was a closet drama designed to be read rather than acted (Milton 1671). Even in genres like opera, where stage action was a feature, wordbooks could enrich that action, often amusingly, for several eighteenth-century writers note that descriptions of the scenery or stage action did not always match what appeared or happened on stage. For example, the Italian text of Rinaldo‘s wordbook calls for the appearance of caged birds, but the parallel English translation states that the birds are to fly around the stage (Rossi 1711, pp. 12, 13). Whichever of these actions actually took place in production, audiences would still have been alive to the alternative in their books.17 Indeed, Rinaldo’s stage directions are extremely detailed throughout, but apparently do not correspond to the reality of its staging, as suggested by the criticisms of Richard Steele in The Spectator (see Stahura 1998). Whether accidentally or deliberately, this opera truly seems to conjure a theatre of the imagination.

Different librettists, then, treated the resource of the wordbook stage directions differently. However, every dramatic oratorio set by Handel contains stage directions in its wordbook, and we should consider those directions a powerful tool in the librettists’ arsenal. We should perhaps emphasize that the absence of staging does not prevent acting. This is still evident today when attending an opera in concert version: the singers stand and perform in formal dress, but can and often do inhabit their characters, use their faces to reflect the emotions they convey through song. For anyone familiar with the plot, the performance comes alive in the imagination. It may also be important to distinguish between an oratorio’s recitative sections and its arias, since the audience probably did not spend the entire time reading the libretto. During the recitatives, listeners can (and arguably would need to) follow the text gradually, as it is usually set without repetition and delivered rather rapidly. In the numerous arias, on the other hand, words are usually repeated multiple times, and the text is delivered far more slowly, so an audience member could probably read the text once and then allow themselves to be carried entirely by the music.

Revisiting Hicks’s assertion that stage directions inevitably disappoint in an oratorio (since they promise what cannot be delivered), we should remind ourselves again of the pleasures of imagination. ‘Heard melodies are sweet, but those unheard are sweeter’, as John Keats put it (Keats 1820, p. 114). The same might be said for images seen and unseen.

2.5. Paratextuality

Bryan White has pointed out (private correspondence) that we could conceptualize the libretto and the setting as two different works, closely related, but distinct. This idea is supported by Zazzo, who explores the idea of wordbooks as ‘paratexts’ with a ‘distinct, parallel identity, which dialogues with, but is not identical to, the actual musical performance’.18 In this framework, the setting (especially if it was never experienced by the reader in person; wordbooks were available at places other than theatres) could form a paratext of sorts for the libretto as read separately; a sort of background myth enriching and informing, but not necessarily circumscribing or dictating, the audience’s reading experience.

In certain cases, this parallel identity is easily seen, for wordbooks frequently included words that would not be performed, or, sometimes, text that the composer never even set. A reader-listener of Messiah, for example, would have been informed in a preface to its wordbook that ‘several things [in the text] are mark’d with a black Line draw down the Margin as omitted in the Performance’ (Jennens 1744, p. 2). In other cases, unset (or at least unsung) words were signaled with double quotation marks, with ‘a Black line drawn in the margin’, or else with an asterisk.19 As Robarts put it, ‘[t]his form of annotation of “silent” lines was standard in wordbooks for musical entertainments: for instance, Motteux’s Hercules. A Masque (c.1697)’ (Robarts 2008, p. 330).20 Thus, the wordbook was not just a reproduction of the sung words, but stood alone in some respects, as a book for reading in whose text the words for listening to are embedded; the libretto is part of the wordbook, but the wordbook contains more than the libretto (see Zazzo 2015, p. 26). And there may also be a more general aesthetic significance to the inclusion of such unset words, since they enrich the experience of the drama, offering new angles on the work as performed and heard, if audiences chose to read them.

2.6. Attitudes to the Libretto

Smith describes oratorio as ‘unstaged, unacted drama, requiring and repaying sustained attention’ (Smith 1995, p. 76). She also states that this type of drama was ‘more essentially verbal than any other form of English-language theatre’, since ‘at no other time did the audience have the words in front of them during the performance’ (Smith 1995, p. 76). The genre was perhaps not quite so unique as this statement makes it seem. In private communication with Zazzo (cited in Zazzo 2015, p. 25), Smith acknowledged that booksellers, and perhaps also theatres, sold the playbooks for ballad operas, which were also usually sung in English. Dramatick operas and masques (usually English-language as well) also used wordbooks, and both odes (usually English) and serenatas (often English) continued to do so in Handel’s time.21 Thus, ‘the audience’ being able to ‘have the words in front of them during the performance’ seems to have been important in many if not most forms of English-language musical theatre.22 Contemporaries even acknowledged this as necessary, due to the difficulty of understanding sung (as opposed to spoken) words, whatever those words’ language might be: defending the performance of Italian-language opera for a largely non-Italian-speaking English public, Burney asked rhetorically, ‘What do we understand when English is singing on our stage without a book?’ (Burney 1789, p. 227).

It might be more accurate to say that Baroque music theatre in general was, at least to some audience members, ‘more essentially verbal’ than twenty-first century scholars often realize. And even if other forms of music theatre did use wordbooks, it is fair to call oratorio even ‘more essentially verbal’ than any of these, since ‘at no other time did the patrons have the words in front of them’ during performance of an ‘unstaged, unacted drama’ (Smith 1995, p. 76): only oratorio consistently asked reader-listeners to imagine stage action based on words that those reader-listeners read silently while experiencing a long music drama in concert performance.23

However, despite all of this attention to communication of verbal detail, and all the labor and centralization of words, eighteenth-century aesthetic attitudes nominally shun libretto writing. As some still do today, many people opined that, as soon as song was attached to poetry, the poetry became trivial; that words penned explicitly for singing must be ‘nonsense’; and that every person with good taste knew that there was no such thing as good music drama (Smith 1997, p. 106). In light of this strange contrast between practice and opinion, and the inconsistency of the expressed opinions themselves, it is difficult to deduce how important the wordbook’s contribution was in the consciousness of the time. Of course, a culture’s asserted values do not always resemble the values revealed in its practice, so perhaps it might be wiser to look at eighteenth-century facts and activities than at the pronouncements of eighteenth-century intellectuals.

Oratorio’s imagined visual dimension was not lost on critics; according to Loughridge (2010), Haydn’s Creation was compared to magic lantern traditions of optical entertainment in the nineteenth century. Some such comparisons were complimentary, others critical, but while they tended to focus on the evocative power of the music rather than the wordbooks, they still reveal a critical engagement with the expressive (and specifically imaginative and visual) power of oratorio as an unstaged genre (Loughridge 2010, pp. 9–10). In Robarts’s view, modern attempts to stage oratorios deprive them of their imaginative power, fixing on a visible stage what was meant to be conjured in the mind’s eye (Robarts 2008, pp. 359–61). Some audiences, of course, might feel that staging makes the oratorios more accessible or enjoyable, so it is perhaps excessively pessimistic to speak of robbery. At the very least, however, we probably need to be as cautious when staging an oratorio as we are when we present an opera in concert. Both presentations alter the recipe fundamental to the drama.

2.7. Directions for Further Research

Many aspects of this topic, and numerous related topics, require further research. Wordbooks allowed audiences to consume the story out of order, frequently included plot summaries, and could even be purchased weeks in advance of the production, so an interested reader-listener would be able to familiarize themselves with the plot and perhaps even the text by the time they arrived at the performance, or at least well in advance of the plot reaching that point or the text being sung.24 These prospective wordbooks (in a manner analogous to modern theatrical trailers, but far more extensive in their deployment of ‘spoiler’ material) at once prioritize and decentralize the narrative dimension of the drama, focusing attention not so much on what happens but on how it happens.

Many wordbooks include typographical cues suggestive of certain performance styles that conflict with how the composer actually sets the words, and this is without even considering how the wordbook’s inclusion of text that was never set (or at least whose setting would not form part of a given performance) informs the processing of the text that was indeed set (or would be performed). Lyrical movements are nearly always introduced with headings like ‘air’, ‘song’, ‘chorus’, ‘duetto’, ‘quartetto’, and so on. Some wordbooks also give headings to texts set in declamatory style, and some of those wordbooks distinguish between simple (continuo-supported) and orchestrally accompanied declamatory singing; the former is generally headed ‘Recitative’, the latter ‘Recitative Accompanied’ (Morell[?] 1757, p. 14). In some wordbooks, including one for Judas Macchabæus (Morell 1746[?]), these headings are the only cues distinguishing between texts sung declamatorily and texts sung lyrically. Usually, however, headings are supplemented by changes of typeface, a longstanding method of indicating the style in which the verse was sung: recitative text tended to be printed in roman type, poetry for lyrical setting was usually in italics.25 In other wordbooks, all the sung text is in roman type, but the typeface for lyrical movements is substantially larger than that for recitative, and, in such cases, lyrical movements almost always have introductory headings, while recitative texts are not (Hamilton 1768[?]). In other wordbooks still, like one for Theodora (Morell 1750), ‘recitative’ headings appear in all caps, the first large, the rest smaller, with the qualifier ‘accompany’d’ in lowercase italics if present (‘RECITATIVE, accompany’d.’); lyrical movements are introduced with headings entirely in large capitals (‘AIR.’). When a recitative is sung by a single character but changes from accompanied to simple or vice versa, the change is generally signaled to reader-listeners not by a new heading but a new line on the page.26

Some wordbooks contain footnotes and attributions of their quotations and paraphrases, and at least some reader-listeners seem to have valued at least these attributions very highly. Smith (2017, p. 176) observes attendees at the première of L’Allegro, Il Penseroso, ed Il Moderato (1740) were disappointed that Milton’s Companion Poems, on which the libretto was based, did not feature in the wordbook (Anonymous 1740, quoted in Smith 2017, p. 176). The libretto is less an adaptation of the Companion Poems than a collage of edited extracts, but reader-listeners still mourned the fact that they could not compare that collage to its source while they read and listened, as evidenced in reviews (Anonymous 1740, quoted in Smith 2017, p. 176):

The two original Poems […] had they been printed entire, and annex’d to the Drama, in a small Character [a small font], it would have been an agreeable Compliment paid to the Audience, had heighten’d the Relish of the Musick, and been no Injustice done to the Fitter of them for the Theatre; whose dramatic Moderato (or Mean between them) can very well bear being compared together.

Although parenthetical, the reference to Moderato as a Mean also suggests close attention to the thematic content of the libretto text: Moderato asserts the superiority of ‘the Golden Mean’ above excesses of mirth or melancholy.

As mentioned, some or most of these features have analogues in later practice; nineteenth-century audiences often purchased pocket scores to follow during performances (Robarts 2008, pp. 353–53; Burrows 2007, p. 113). And modern audiences are frequently given programs containing the words to be sung. But it seems interesting to investigate the use of such elements not as optional extras but near-essential co-products of the original experience (Twomey 2020, pp. 85–197). Baroque wordbooks were not designed to demystify historically remote music, but to draw audiences deeper into the music-drama of their contemporaries.

Moreover, since many arias’ instrumental introductions are based on their first vocal phrase, an exciting dimension of predictive play arises if the audience member is reading the text while listening to the music. It would be tempting for an audience-member to guess how the instrumental introduction will shortly be reiterated by the voice, and how it will set the text; they might even be tempted to sing the predicted setting to themselves. And the composer could make fascinating use of that predictive imagination, satisfying such expectations, subverting them, or mixing satisfaction with subversion in various ratios (Twomey 2020, pp. 85–197).

3. Conclusions

Returning to the questions posed at the beginning of this article, it is clear that wordbooks had, or at least had potential to have, profound influence on the processing of the performed work. By providing evocative stage directions, they invited the imaginative construction of images not physically present on stage, and would often be the only way that the audience could follow the plot. The presence of unset text offers a fascinating framing for the text actually performed, and the ready availability of the complete text decouples the narrative from a need for linearity, centralizing ‘spoilers’ and drawing attention away from what happens in the drama to focus upon how things happen.

Undoubtable, not every member of the audience had a wordbook to read at every performance. Perhaps they were not aware of the wordbook’s importance, or could not afford one, or were illiterate; perhaps the theatre ran out of copies, or the audience member simply lost their place in the text. It is also entirely possible that many audience members simply did not care about reading wordbooks. This article posits a specific type of musico-linguistic connoisseurship, but it makes no claim that everyone was a connoisseur. However, a resource not universally used is not necessarily an irrelevant or universally unused resource. Surely not every modern listener reads the liner notes of every album to which they listen, but that does not make the practice of producing or reading such liner notes any less interesting.

To point out the potential enrichment that wordbooks can bring to unstaged oratorio performance is emphatically not to call for an end to oratorio staging, for mandatory distribution of wordbooks, or for ‘inauthentic’ listeners and performers to be beaten with yet another stick. At its conclusion, however, this article hopes to have shown that oratorio, even in its unstaged form, does indeed incorporate, in some cases even depend on, stage action, movement, scenery, and spectacle. All the visuality of opera is there. But, as the Goons often reminded their own listeners about the visuality of radio, ‘[i]t’s all in the mind, you know’ (Milligan and Stevens 1957).