JOHN WEBSTER 
 
 and the 
 
 ELIZABETHAN DRAMA 
 
 liy 
 
 RUPERT BROOKE 
 
1 
 
 LIBRARY 
 
 UNIVERSITY OF 
 CALIFORNIA 
 
JOHN WEBSTER and the 
 ELIZABETHAN DRAMA 
 
The 
 
 COLLECTED POEMS 
 
 of 
 
 RUPERT BROOKE 
 
 WITH A 
 
 PHOTOGRAVURE PORTRAIT 
 0/ the AUTHOR 
 
 Cloth, $1.25 net 
 
 Leather, $2.00 net 
 
 "It is packed with the stufif of which poetry 
 is made: vivid imagination, the phrase that 
 leaps to life, youth, music, and the ecstasy 
 born of their joy when genius keeps them com- 
 pany." — The Outlook. 
 
 JOHN LANE COMPANY 
 PUBLISHERS NEW YORK 
 
JOHN WEBSTER and the 
 
 ELIZABETHAN DRAMA 
 
 By RUPERT BROOKE 
 
 JOHN LANE COMPANY 
 NEW YORK .-. .-. .-. MCMXVI 
 
Copyright, 1916, 
 By John Lane Company 
 
 Press of 
 
 J. J. Little & Ives Company 
 
 New York, U.S.A. 
 

 V-; 
 
 
 NOTE 
 
 This hook was written in 1911-12, and was the 
 ^dissertation' with which the author won his Fellow- 
 ship at King's College, Cambridge, in 1913. 
 
 The page-references are to Dyce's one-volume 
 
 edition. 
 
 E.M. 
 
 4 \J,*<^ 
 
C^"' *'"' 
 
PREFACE 
 
 I HAVE tried to write a small book about John 
 Webster. That is to say, I have tried to say 
 the truth about him, as much of it as is 
 necessary to enable anyone who reads him to 
 understand him. I have not tried to explain him 
 entirely to anyone who has not read him, though 
 I hope that any person in that condition may 
 get a rough idea of him from this book. 
 
 I have tried to explain Webster for a reader, 
 but not to explain him away. So I have endeav- 
 oured to keep to my own province, and not to 
 trespass on ground reserved for worthier feet — 
 Webster's. I conceive that there is much that he 
 can explain better than I. So I have, at least, 
 abstained from paraphrasing. 
 
 To explain Webster's writings it is first neces- 
 sary to determine what he wrote, and also such 
 smaller questions as when he wrote it, and how 
 he came to write it. Such questions, the ques- 
 tions of "scientific" literary criticism, I deal with 
 in the Appendices. I have taken some care to 
 get the most probable answers in each case; for 
 there is such a lot of bad logic and fudging on 
 
 vii 
 
viii PREFACE 
 
 such points in modern literary science, that one 
 always has to go over the whole ground com- 
 pletely for oneself. 
 
 When these points are settled, with as much 
 certitude as possible, there are still other points 
 on which it is necessary to have right opinions in 
 order to understand Webster. One must know 
 what a play is; one must laiow how the Eliza- 
 bethan drama arose; and one must know what 
 the Elizabethan drama was. I have given a 
 chapter to each of these points; not pretending 
 to cover the whole ground, or to do the work of 
 a whole book; but endeavouring to correct some 
 of the more misleading wrong ideas, and to hint 
 at some of the more important right ones. These 
 chapters, of course, though nominally not about 
 Webster, should be even more important to any 
 understanding of him than the Appendices. And 
 I have given two long chapters to the more 
 direct consideration of what Webster wrote, and 
 what its more usual characteristics are. 
 
 The Bibliography is, I think, fairly complete 
 with regard to Webster. I did not think it neces- 
 sary to make a bibliography of books on the 
 wider subjects. 
 
 It may seem, in some cases, as if I contra- 
 dicted myself in different parts of the book; as, 
 for instance, when I say that it is impossible to 
 
PREFACE ix 
 
 understand a play wholly from the text, and later 
 seem to believe that I do understand plays wholly 
 from the text. I think I have not really contra- 
 dicted myself. Part of the business of the earlier 
 chapters is to prevent the necessity of continually 
 repeated qualifications throughout the work. To 
 express my exact meaning on each occasion 
 would have meant covering the page with 
 "in so far as it is possible's," and "I think's," and 
 "possibly's," and "perhaps's"; which makes the 
 style feeble and muffles the idea. I have, per- 
 haps, gone too far in this direction already. 
 
CONTENTS 
 
 Preface ....••* 
 
 CHAPTEB 
 
 I. The Theatre .... 
 
 II. The Origins of Elizabethan Drama 
 
 III. The Elizabethan Drama . 
 
 »^IV. John Webster .... 
 
 i V, Some Characteristics of Webster 
 
 PAGE 
 
 vii 
 
 15 
 
 38 
 
 62 
 
 84 
 
 123 
 
 Appendices 
 
 A. The Authorship of the later Appius and Virginia l65 
 
 ^^. Miscellaneous 
 
 C. Sir Thomas Wyatt 
 
 D. Westward Ho and Northrvard Ho 
 
 E. The Malcontent . 
 
 F. The White Devil . 
 
 G. The Duchess of Malfi . 
 H. A Monumental Column 
 
 I. The DeviVs Law-case . 
 J. A Cure for a Cuckold . 
 
 1/ Bibliography .... 
 
 211 
 214 
 222 
 234 
 237 
 246 
 254 
 255 
 260 
 
 277 
 
JOHN WEBSTER and the 
 ELIZABETHAN DRAMA 
 
JOHN WEBSTER 
 
 Chapter I 
 
 THE THEATRE 
 
 Anyone who has read, with any alertness, more 
 than a little of the mass of critical and editorial 
 comments, whether of the last three or of the last 
 three hundred years, upon Elizabethan plays, 
 must often have felt a helpless and bewildered 
 irritation at the absence of any standard or uni- 
 form grounds of judgment; both in the critics, 
 and, on inspection, in himself. This is not the 
 place to attempt to lay a deep aesthetic founda- 
 tion; but, I think, it will be useful to try to fix 
 the meanings of certain words and phrases, and 
 to give a provisional answer to some of the more 
 important questions. 
 
 "What is Art?" is a question which most 
 writers on subjects connected with literature, 
 painting, plays, music, society, or life, are ready 
 with an equal cheerfulness to ask or to answer. 
 They may be right; but to me they seem to make 
 
 15 
 
16 JOHN WEBSTER 
 
 a gigantic, unconscious, and probably unjustifi- 
 able assumption. It is quite doubtful, and it is 
 nowadays continually more doubted, whether the 
 word "Art" has properly any meaning at all. 
 But it has so obsessed men's minds, that they 
 start with an inevitable tendency to believe that 
 it has a meaning. In the same way, those who 
 believe in Art are generally inclined to believe in 
 a single object at which all Art, that is to say all 
 the arts, aim: Beauty. It may turn out to be 
 true that both Art and Beauty are real and 
 useful names; but the attitude of mind that as- 
 sumes that they are is deplorable. The most 
 honest and most hopeful course to pursue, is to 
 say that there are certain kinds of human activity 
 which seem to hang together in classes, such as 
 reading books, hearing music, seeing pictures; 
 and to examine our states of mind while we fol- 
 low these pursuits, to see how far they are of 
 one kind in each "art," and in all, and whether 
 all successful works of art do seem to us to have 
 some quality in common which can be called 
 Beauty. 
 
 The situation seems to me as if men had agreed 
 to say "The emotions caused in human beings 
 by pins, walking-sticks, feathers, and crowbars, 
 acting through the tactile sense, are all of one 
 unique kind. It is called Grumph. Pins, etc., 
 
THE THEATRE 17 
 
 are called the grumphs. Grumph is one of the 
 holiest things in this melancholy world," and so 
 forth. And soon they'd say, "But, philosophi- 
 cally, what is Grumph?" Then they'd argue. 
 They would come to some conclusion which, as 
 you cannot tickle with a crowbar, would pre- 
 clude tickling with feathers; and they would ex- 
 communicate all those who used feathers for 
 tickling with the formula, "That is not 
 Grumph!" They would write Treatises on any 
 one grumph, on the "Pin-grumph," say, care- 
 fully keeping in mind all the time that what they 
 said would have to be more or less true of the 
 other grumphs too. Some would lay great im- 
 portance on the fact that, as you were tickled 
 with feathers, you were, in a way, also tickled by 
 being beaten with a walking-stick. Others 
 would discover the ferule of the pin, and the 
 quill, shaft, and two vanes of barbs of the crow- 
 bar. An Oxford don would arise to declare that 
 all grumph continually approximated to the con- 
 dition of pins. . . . 
 
 I have put the affair, as I see it, in a figure, 
 and with other names, in order to show its un- 
 reason more clearly, and far more shortly, than 
 is possible if the prejudice-clad and elusive word 
 "Art" is used. In either case, the sensible reply 
 to it all is, "We have sticks and pins, plays and 
 
18 JOHN WEBSTER 
 
 poems. These we know. These are, as certainly 
 as anything is, real classes of things. Begin from 
 them, and from the emotions they move. And 
 see if thence you climb upwards to Grumph, to 
 Art.'' 
 
 This attitude does, directly or indirectly, shut 
 out various bands of ideas and thinkers ; my ob- 
 jections to each of which I could state at length. 
 A short enumeration of these tendencies of mind 
 in viewing questions of "Art" may hint why, 
 psychologically at any rate, they seem to me non- 
 starters. In the first place, I do not admit the 
 claims of anyone who says, "There is such a 
 thing as Beauty, because when a man says, 
 'This is beautiful,' he does not mean, 'This is 
 lovely,' or, 'This provokes the cosmic emotion.' 
 There is such a thing as Art; because the sen- 
 tence: 'Pictures, Poetry, Music, etc., are Art,' 
 is not the same as 'Pictures, Poetry, Music, etc., 
 are Music, Poetry, Pictures k. t. x.' " I am not 
 concerned with what men may mean. They fre- 
 quently mean and have meant the most astound- 
 ing things. It is, possibly, true that when men 
 say, "This is beautiful," they do not mean "This 
 is lovely." They may mean that the esthetic 
 emotion exists. My only comments are that it 
 does not follow that the aesthetic emotion does 
 
THE THEATRE 19 
 
 exist, and that, as a matter of fact, they are 
 wrong. 
 
 But the only way to prove them right or wrong 
 is by introspection into our states of mind when 
 we hear music or see pictures. 
 
 It has been acutely said that, in philosophy, 
 it is important to give the right answers, but even 
 more important to ask the right questions. So 
 here. Better than to ask "What is Art?" is it 
 to ask "What do you feel before this picture?" 
 "Before that picture?" "Is there anything com- 
 mon between your feelings in these two cases?" 
 "What do you feel in hearing this, and that, piece 
 of music?" "Is there anything common?" and 
 then, "Is there anything common between what 
 you feel before all these pictures and what you 
 feel in hearing all this music?" "And if so, what 
 is it?" "Is it important?" One of the perils 
 attending on those who ask the first question is 
 that they tend, as all men do, to find what they 
 are looking for: a common quality in Art. And 
 also that they tend to exalt what they discover 
 for this quality, above the others that are to be 
 found in any of the arts. People who start in 
 this way are apt to be, practically, a most in- 
 tolerable nuisance both to critics and to artists; 
 whether it is Art or any one art that they would 
 tie to their rule. Art is Pattern; and a novel 
 
20 JOHN WEBSTER 
 
 that lacks "pattern" is not Art, and therefore 
 bad. Art is the perception of the individual 
 case; so morality plays are illegitimate. Art is 
 the emphasising of the generality; so Hamlet, 
 except in so far as the hero represents all neuro- 
 paths, is a perverse and downward path from 
 the moralities. Art must be moral; so Shake- 
 speare's sonnets are what Hallam thought them. 
 Art has no connection with morality ; so Paradise 
 Lost and Pilgrim's Progress are, artistically, 
 worthless. A play must display a "develop- 
 ment,'' a tragedy must involve a conflict ; music 
 must have a tune ; a picture may not tell a story. 
 . . . The list of these perilous and presumptu- 
 ous a priori limitations could go on for ever. Of 
 the wrong ways of approaching the subject of 
 "Art," or even of any one art, this is the worst 
 because it is the most harmful. 
 
 But there are other ways in which precon- 
 ceptions and assumptions about the thing to be 
 looked for mislead, in the consideration of Art. 
 Croce rather naively begins by noting that "aes- 
 thetic" has been used both for questions of Art 
 and, in general and in accordance with its deriva- 
 tion, for perception. So he sets out to discover 
 what meaning it can really have, to apply to 
 both. He takes it for the one necessary condi- 
 tion a true answer about "^Esthetics" must sat- 
 
THE THEATRE 21 
 
 isfy, that it shall explain how Art and Percep- 
 tion are both included. Having found such an 
 explanation, he is satisfied. 
 
 To take a different side, most of the uphold- 
 ers of the Einfiihlungsdsthetik seem to have 
 founded their view on the experiences of the spec- 
 tator of certain visual arts, especially painting 
 or architecture. In so far, it is valuable. But 
 when it is contorted to cover the other arts, the 
 result is ludicrous. So those who accept the 
 Nacherleben theory, would appear to be extend- 
 ing what is probably true about drama to spheres 
 where it is desperately irrelevant. 
 
 It is said that the figure of Helen, whom men 
 have so eagerly followed and sought, was a 
 phantasm, covered by which there lurked, in fact, 
 a knot of mercantile interests of Greece and the 
 Hellespont and the Black Sea; even as, some 
 claim, men who have died for the love of Eng- 
 land, or Germany, or Italy, have, in reality, 
 only given themselves for a few rich people. Art 
 and Beauty have proved such delusive Helens. 
 It is an extraordinary crowd, pouring along di- 
 verse roads, that has followed them. The on- 
 looker is moved to amazement and derision. Ros- 
 setti's "View Halloo!" was less lonely than he 
 dreamt. More than all illusory goddesses has 
 My Lady Beauty been chased or stalked, as a 
 
22 JOHN WEBSTER 
 
 rule passionately, often irretrievably, "in what 
 fond flight, how many ways and days!" The 
 ingenuity of the chase has been stupendous. 
 
 "They sought her with thimbles^ they sought her with care; 
 They pursued her with forks and hope." 
 
 The thimble of an a priori generalisation has 
 not closed down on My Lady, nor the fork of 
 Dialectic impaled her. For the quest was vain 
 from the beginning. It is that conviction that 
 enables me so cursorily to leave such knight-er- 
 rants to their task — of "bounding along on the 
 tip of their tail" or "still clutching the inviola- 
 ble shade," according to the way you regard 
 them. We had best cultivate our gardens of 
 the arts. Then we may turn round one day to 
 discover Beauty at our elbow — if she exists at 
 all. If she doesn't, we shall at least have learnt 
 horticulture. 
 
 I can descend, then, with a clear conscience 
 to occupy myself with the single plots of ground 
 called Drama and Tragedy. But first I must 
 deal with two other ways of approaching the 
 question of the arts — for the arts, as human 
 activities, can be classed together, even though 
 there be no such obvious similarity discernible 
 in the states of mind they produce, no "aesthetic 
 emotion." There are some who would view it 
 
THE THEATRE 23 
 
 all from the point of view of the artist. "Art," 
 they say, "is primarily a creative function of the 
 artist; other people may profit, afterwards, if it 
 so happens. Cricket is a game played by twen- 
 ty-two men, under certain rules: which may or 
 may not be watched by a crowd. This is true, 
 even though the game would not have been 
 played but for the crowd. Art is no more to be 
 explained in terms of the chance result on the 
 spectators than cricket is to be explained in 
 terms of the feelings of the crowd. Art is an 
 amazing creative experience of the artist: what 
 happens to the result of his travail is neither here 
 nor there. A good picture is one in the creating 
 of which the artist had a good state of mind. 
 And the utmost a spectator can hope for is to 
 approximate, in beholding a work of art, to the 
 state of mind the artist had in creating it." 
 
 The last sentence, perhaps, expresses a view 
 that need not logically go with the foregoing 
 belief. For the whole position, I do not think 
 it can, ultimately, be refuted. It becomes a 
 question of words, or of the point of view. From 
 where I stand, I seem to see certain activities, 
 and I consider them according to the aspect that 
 seems to me most important. If another man 
 views and describes them from behind, I can only 
 lament it. There are things to be said against 
 
24 JOHN WEBSTER 
 
 him. Certainly, if importance is to weigh in the 
 matter, the effects on the audience are more im- 
 portant than the state of the artist. He could, 
 cogently, answer that corn is corn, though the 
 most important thing about it is that it goes to 
 make bread. A greater difficulty is the extraor- 
 dinary variety of experience of the creative 
 artist. Blake thought he was taking down his 
 writings from the dictation of an angel. Some 
 writers solemnly think their things out. Others 
 are "inspired"; or proceed almost by automatic 
 writing. Some are highly excited and irrespon- 
 sible; others detached, cynical, and calculating. 
 Many artists, it would seem, are never aware 
 of their work of art as a whole, but build it up, 
 patching and revising in little pieces. A play 
 by Beaumont and Fletcher, with the scenes ap- 
 portioned out, would be difficult to judge by this 
 creative theory. Certainly, if you take the case 
 of a dancer, who can never quite see herself danc- 
 ing, it seems clear that the important whole con- 
 nected with this activity is in the state of mind 
 of the spectator. 
 
 Another common tendency, a fatal and ridicu- 
 lous one, is that of the historical school. Both 
 the psychology of the artist and the history of 
 the arts are interesting, and may be valuable, 
 topics of investigation. But it should be clearly 
 
THE THEATRE 25 
 
 recognised that the history of the forms of the 
 arts has no direct connection with the arts as they 
 are. Football originated in a religious ritual; 
 but it is not, necessarily, religious. The cooking 
 of roast pork arose from the burning of a house ; 
 but he would be a foolish gastronomist who, in 
 considering cooking, laid great emphasis on the 
 fundamental element of arson in that art. So 
 there are some who say that the arts originated in 
 a need to let off the superfluous energies of man, 
 not needed to further or secure his livelihood; 
 and therefore are essentially of the nature of 
 play. Others declare that the sexual instinct 
 was at the bottom of the beginnings of the arts, 
 and that all Art is, fundamentally, sexuality. 
 Others again would, for similar reasons, find it 
 a religious activity. To all such we can only 
 reply, "If your historical analysis is true, it is 
 indeed a wonderful world in which we live; but 
 now, in 1912, poetry and football are not sex 
 or religion; they are poetry and football." 
 
 There are theatres; places where you see 
 things. The things you see there generally try 
 to represent or imitate reality, and are frequently 
 accompanied by words, in which cases they are 
 called "plays." One of the first and most im- 
 portant distinctions between plays, music, and 
 poetry on the one hand, and pictures and sculp- 
 
26 JOHN WEBSTER 
 
 ture on the other, is that the element of duration 
 enters into the first group. There is no especial 
 point in a picture at which you begin or end look- 
 ing at it; no fixed order of sensations. There is 
 just the picture. But the order of sensations 
 which a play should arouse in you is fixed be- 
 forehand, and essential. This fact of duration 
 gives theatrical art two features. It can arouse 
 all the emotions that can be got through the con- 
 secution of events; and it can employ the suc- 
 cession of emotions in the mind. Both these 
 are important. Take the latter first. It is obvi- 
 ous that, though he may demand certain knowl- 
 edge in the spectator before the beginning of the 
 play, the artist cannot demand any definite state 
 of mind. He can only claim to be presented with 
 an expectant and fairly blank normal mind. 
 After that he is responsible. And at any moment 
 during the play, his choice of the emotions to 
 arouse is conditioned by the emotions already 
 aroused. Each situation must be planned, each 
 line written, with regard to the effect of what has 
 gone before, not only logically, but psychologic- 
 ally, on the audience. The continuity of the play 
 must be an emotional continuity, even more than 
 a rational one: not necessarily, of course, the 
 same emotion continuously, but necessarily har- 
 monious ones, I do not mean to suggest that the 
 
THE THEATRE 27 
 
 spectator of a play experiences a number of defi- 
 nite emotions, one at a time, each lasting three 
 seconds, consecutive. His state of mind is com- 
 plex; and while some perceptions or emotions 
 flash with infinite swiftness through it, others last 
 and colour the contents of subsequent states of 
 mind for some time. It is these last that are 
 most important, but the whole mental and emo- 
 tional experience has a cumulative effect. It is 
 as if a stream of water of various heat was trick- 
 ling through a basin. The heat of the water in 
 the basin at any moment would be affected by 
 the heat of the basin, which in turn would be 
 a result of the past heats of all the water that 
 had gone through before. Only, heat is simple, 
 and the succession of emotions and sensations is 
 manifold and complex. The merit and kind of 
 the play, in a sense the play itself, lie in the whole 
 curve of these states of mind. That is the most 
 important thing about plays, to which every- 
 thing, ultimately, must be referred. I can more 
 easily imagine a play good in which all the char- 
 acters of the first four acts vanished, and entirely 
 new ones came on in the fifth, with an entirely 
 new plot, so long as the emotions aroused were 
 harmonious, than one in which the successive 
 states of mind clashed. 
 
 What a man generally refers to when he 
 
28 JOHN WEBSTER 
 
 speaks of a play, and of the goodness and quali- 
 ties of it, is a memory of this succession of states 
 of mind, a kind of foreshortened view of it, an 
 emotional precis or summary. A good critic is 
 he who can both feel a play perfectly at the time, 
 and sum up its particular taste and intensity 
 perfectly, for his own reference, in this retro- 
 spective summary. The process of summarising 
 a play thus involves the abstraction of various, 
 more or less common elements of the successive 
 states of mind the play produces, and the con- 
 cocting them into one imagined taste or state of 
 mind, "the play." All these summaries are of 
 something the same kind; so the habit of think- 
 ing of plays thus leads men to think that there 
 is some common quality in all of them — at least, 
 in all serious ones — "beauty" and a common "ses- 
 thetic emotion" always in the mind of all spec- 
 tators of plays. I believe that honest introspec- 
 tion of one's states of mind during a play, will 
 show that there is no one quality one can call 
 "beauty" in all successful serious plays. If there 
 is any meaning at all in the word "beauty," my 
 emotion at lago's temptation of Othello, or 
 Lear's "Prithee, undo this button," is in no way 
 a consciousness of beauty; and though there is, 
 perhaps, something in my state of mind — the 
 shape of it, so to speak — which is the same when 
 
THE THEATRE 29 
 
 I watch any tragedy, it is only due, I think, to 
 the fact that all tragedies I know have a certain 
 common quality of being partly like life; I do 
 not find this something in my mind when I am 
 watching pure dancing. 
 
 A play is good in proportion as the states of 
 mind during the witnessing of it are, in sum, 
 good. The good of these states of mind is, in 
 practice, very much dependent on the pleasur- 
 ableness of them, and proportionate to it. Much 
 more so than in real life, where the consciousness 
 of virtue makes some unpleasant states good. 
 But pleasure is not a perfect criterion of good, 
 even in the theatre. For a performance that pro- 
 vokes lust would move pleasant states of mind, 
 but not good ones. 
 
 If this is granted, the difficulty is: in whom 
 is a play to move good states of mind, in order 
 to be called good? Obviously, not only in me. 
 A play in Russian might be very good, and yet 
 only bore me, because I couldn't understand it. 
 On the other hand, I do not think it fair to call 
 a play good which can be understood by nobody 
 but the author. Everybody is familiar, in the 
 realm of literature, with the writer who is im- 
 mensely pleased with his own poem because of 
 the emotions it evokes in him. The phrase "the 
 sun is setting" recalls to him the purple and 
 
30 JOHN WEBSTER 
 
 green glory that moved him to this inadequate 
 expression. But it will not affect anyone else in 
 the same way, so we rightly refuse to call the 
 poem good. Obscurity in an author is, ulti- 
 mately, a fault. A family of my acquaintance 
 uses a private and peculiar synonjTn of their 
 own childish invention for "hand," the word 
 "nopen." ^ If one of them wrote a poem con- 
 taining this word, it would affect him very much, 
 because of the aura of associations around it. 
 But the rest of the world would find it mean- 
 ingless. It would not be a good poem. One is 
 reduced to saying that a good play means a play 
 that would be likely to stir good states of mind 
 in an intelligent man of the same nation, class, 
 and century as the author. It follows that a 
 good Elizabethan play is a play that would have 
 been good in Elizabethan times; and not a play 
 that is good to us, with our different ideas. The 
 two categories coincide to a great extent. But 
 their differences are important. 
 
 And it follows that all those literary qualities 
 that answer to patine in works of art — quaint- 
 ness, old-fashionedness, interest as illustrating a 
 bygone age — are irrelevant. I had rather read 
 an interesting book originally worthless, than a 
 fine poem in a language I cannot understand. 
 
 * Because it opens. 
 
THE THEATRE 31 
 
 But it would be misleading to call the former a 
 better book. 
 
 Whether the states of mind produced by a 
 play were good or not, must be decided by intro- 
 spection. The object of most critical enquiries 
 is to discover what sort of effect different things 
 in the theatre have on these states of mind. It 
 is obvious if one examines one's consciousness 
 during a play, that several different classes of 
 object fill and move it. There is sound. Music, 
 or the mere melody of words, impresses and 
 pleases. There is the further literary pleasure of 
 the language, apart from the mere sense; and 
 sometimes there is metre. There is movement, 
 varying from absolute dancing to mere imitation 
 of life. There is, in most theatrical perform- 
 ances, the story. And there is the realism of the 
 piece ; i.e. its value as impressing us with the sense 
 of its reality. 
 
 If we exclude pure dancing, all performances 
 in theatres have some value as connected with 
 reality. To discover what it is, one has to con- 
 sider one of the widest and most important psy- 
 chological questions connected with the theatre, 
 the question of convention. 
 
 To say that one feels the reality of an ordinary 
 play without believing it, is a fairly accurate de- 
 scription of one's attitude. It would be better 
 
32 JOHN WEBSTER 
 
 to put it in this way: the feeling of reality, the 
 emotion of conviction, of faith, is a purely psy- 
 chological one. It is this that plays aim at pro- 
 ducing. It is not the same emotion we have in 
 real life. In real life one does not feel "He is 
 really there, talking to me!" One takes it for 
 granted. He is there. This is also present to 
 some degree when one is witnessing a play, but 
 it is the negative and less valuable side of the 
 emotion. The former, the positive feeling of 
 reality, does not tend to result in action. The 
 latter does permit of various emotions resulting 
 in action. So there has to be a permanent inhi- 
 bition of such action ; or, to put it in another way, 
 you accept the convention of the actors, the 
 absent fourth wall (on the modern stage), and 
 so on. It was in the want of this inhibition that 
 the wrongness of that Italian's attitude lay, who, 
 at a performance of Hamlet^ was so wrought 
 upon that he rose from his place in the pit, and 
 shot Claudius. Many find it difficult to under- 
 stand the attitude of the human mind about such 
 convention. They either say, "Absence of scen- 
 ery destroys the illusion," or "You must know 
 it isn't true." The accepting of a convention 
 means that one says, "Suppose Romans talked 
 
 English blank verse, then " and gives oneself 
 
 to the play; or, to put it another way, one puts 
 
THE THEATRE 33 
 
 a lid on one's knowledge that Romans didn't talk 
 English blank verse. Ignorant of that, one can 
 believe the rest. 
 
 This is one of the most natural and deep- 
 rooted instincts in men. We do not want illu- 
 sion; we only ask that conventions should be 
 made and kept. But it is important that they 
 should be kept. The artist can make any amount 
 of conventions; but, once made, he must not 
 break them. It is obvious in children. A grown- 
 up can say, "Suppose you are a hen, and she is 
 a steam-roller, and I am the King of Portugal," 
 and they will carry the play out with entire ac- 
 ceptance of this, absolute appreciation of the 
 drama ensuing. But if the grown-up breaks 
 from his regal speech and behaviour a moment 
 to address a remark, in his own person, to some 
 outsider or to the steam-roller in its private exist- 
 ence, the grief and dismay of the children is 
 prodigious and unexpected. Observation or 
 memory will assure one that their pain is purely 
 aesthetic. It is what we feel when a dramatist 
 breaks or misuses one of the conventions. 
 
 The artist's business, then, is to make these 
 various conventions, and, within them, to impress 
 the spectator as much as possible with the sense 
 of reality. There are many ways of doing this; 
 realism in any one one branch — in the chain of 
 
34 JOHN WEBSTER 
 
 events, in the gestures of the actors, in the style 
 of speech, in the truth to life of the characters, 
 or in the scenery — will do to start the feehng of 
 reality, and it will then gather force from the 
 general power of the play. Or there are unreal- 
 istic ways of impressing the spectator with real- 
 ity, through mere literary or theatrical power. 
 It is to be noticed that in some of these things, 
 realism means breaking a convention and setting 
 up a more realistic one, and is consequently com- 
 parative. With speech, for example, realism 
 means more realistic speech than one is accus- 
 tomed to. Robertson's Caste was realistic in this 
 direction, in its day. When we had got used to 
 that, Mr. Shaw's plays, with their more natural- 
 istic speech, appeared, and seemed to us more 
 realistic. They, in their turn, ring now old- 
 fashioned by the side of more modern plays, the 
 dialogue of which seems to us, for a time, start- 
 lingly and triumphantly like real life. 
 
 If one keeps in mind the fact that the ultimate 
 classification of plays, for aesthetic purposes, 
 must be by the general tone of the states of mind 
 they evoke, the endeavour to distinguish Trag- 
 edy from Comedy, and to define Tragedy, by 
 subject-matter, appears rather misleading. 
 Tragedy may have to have a "hero," it may in- 
 volve death, it may require a conflict. All we 
 
THE THEATRE 35 
 
 know is that, in the two or three varieties of 
 Tragedy we are acquainted with that have hith- 
 erto been evolved, these things are generally 
 present. The duty of critics is rather to decide 
 how far it is probable that a play with a hero will 
 evoke deeper "Tragic" feeling than a play with- 
 out one, and such half-technical and quantita- 
 tive questions. 
 
 The emotions of a spectator are produced in 
 various ways, and through the two channels of 
 the eye and ear. Performances can mix their 
 appeals through these channels in any propor- 
 tion. Pantomime can appeal, very powerfully, 
 through the eye alone. A blind man could get 
 a great deal of enjoyment out of some plays. 
 But honest introspection will convince anyone 
 that a very large part of the appeal made by a 
 performance of the kind of play Hamlet or The 
 Duchess of Malfi is, comes through the eye. 
 Would one rather be bhnd or deaf at such a 
 performance? It is a comprehensible and com- 
 mon, but dangerous fault, to over-emphasise the 
 importance of the printed text to the whole play. 
 It is true that the romantic halo and additions of 
 beauty to the general lines of the play, came, in 
 Elizabethan plays, very little in the things you 
 could look at; almost entirely in the words. But 
 the story itself was told visually as well as audi- 
 
36 JOHN WEBSTER 
 
 bly. The Elizabethans were above all men of 
 the theatre, and planned performances. It is 
 important always to keep this in mind when read- 
 ing their "plays," always to be trying to visualise 
 the whole performance from the text, and to 
 judge it so, and always to look with suspicion on 
 those who judge the text as literature. It may 
 be good literature, sometimes ; but it was not pri- 
 marily that. To judge The Duchess of 3Ialfi 
 from the book of the words which we happen to 
 possess is a little like judging a great picture by 
 a good photograph of it. The general plan is 
 given you, and you see all the lines, and shapes, 
 and shading; and you have to supply the colour 
 by an effort of the imagination. Much genuine 
 aesthetic pleasure can be got from this; but no 
 one would be so rash as to assume that, after 
 that, he knew the picture. With plays, people 
 are more presumptuous. But an honest man 
 will sadly have to acknowledge that, in the text, 
 we have only the material for a rough, partial, 
 and hesitating appreciation of The Duchess of 
 Malfi; and that this is the truer because it is an 
 Elizabethan play, that is to say, it is written in 
 a language somewhat different from ours, and 
 pronounced differently too, and it was per- 
 formed in conditions we do not completely know 
 and cannot at all realise. It was composed for 
 
THE THEATRE 87 
 
 an audience accustomed to the platform stage 
 and no scenery; which we can never be. It was 
 composed for the stage, and we judge it as liter- 
 ature ; we are only readers. It is right enough to 
 attempt to realise imaginatively Elizabethan 
 plays as plays. It is right enough to admire 
 their great literary merits and their rather acci- 
 dental power as study-drama. But, after all, 
 we have only the text — and that a not always 
 trustworthy one — one factor of several in the 
 play, a residue, fragments of the whole. We are 
 like men who possess sweet-smelling shards of 
 a jar which once held perfumes, and know how 
 fragrant it must have been; but the jar is broken, 
 and the perfumes lost. 
 
Chapter II 
 
 THE ORIGINS OF ELIZABETHAN 
 DRAMA 
 
 It needs the imaginative sympathy of a good 
 anthropologist to understand the real nature of 
 the various progenitors of the Elizabethan 
 drama; and it needs the intuition of a good psy- 
 chologist to interpret it. Luckily much of the 
 outer history, names, dates, and facts, together 
 with a good deal of understanding explanation, 
 has been given us by such writers as Professor 
 Creizenach, and, above all, by Mr. Chambers. 
 Subsequent works, such as The Cambridge His- 
 tory of English Literature, merely follow on 
 his lines, sometimes slightly varying relative im- 
 portances, nothing more. But as one reads the 
 array of facts and the brilliantly powerful gen- 
 eralisations and inductions of JNIr. Chambers, or 
 the patient condensations of his successors, it is 
 impossible not to feel the full sea of scepticism. 
 Where we have records, do we really understand ? 
 It is hard enough, four-fifths of the books now 
 written on them witness, not to be wholly out of 
 
 38 
 
ORIGINS OF DRAMA 39 
 
 touch with the Elizabethans themselves. But 
 they are our brothers and fathers. These others, 
 these white-faced savages who seem to beckon 
 and move in the fog of the Middle Ages or the 
 deeper night behind — what have they to do with 
 us? A surface likeness of name and tongue will 
 not hide their foreignness. Their hearts are dif- 
 ferent, and distant from ours. They live in an- 
 other universe. The unconscious worshippers 
 of a vegetation-god, the audience of a scoj), the 
 spectators of a miracle-play — what was really in 
 their minds? We triumphantly know that the 
 Feast of Fools was celebrated at Tournai on the 
 eve of Holy Innocents, 1498, that an interlude 
 was given at King's Lynn on Corpus Christi 
 1385, that the processional religious drama was 
 acted on "pageants," and so forth. But what 
 were the people thinking, as the waggons rolled 
 by or the actors came out? How like was it to 
 an Elizabethan's feeling as he watched The 
 Tragedy of Byron? or to ours when we see The 
 Importance of Being Earnest? It is absurd to 
 pretend we know. 
 
 Such are the misgivings with which the honest 
 student looks back on "the origins of the 
 drama." He can pretend he sees how the "plat- 
 form-stage" arose, and passed into the "picture- 
 stage"; he can cheat himself into believing he 
 
40 JOHN WEBSTER 
 
 has established the generations of an Enghsh 
 dramatic form; but what, in our time and race, 
 is the history of those comphcated states of mind 
 the witnessing of Hamlet breeds in us — that he 
 dare only wonder. 
 
 If he looks beyond the Middle Ages he finds 
 at first more familiar things. Seneca's plays fall 
 recognisable on his modern hearing ; and if those 
 were never on the stage, other tragedies and 
 farces which we could, it is imaginable, under- 
 stand, if not applaud, held the Roman ear. And 
 the modern eye greets even more gladly finer, 
 less recorded, performances. The best taste in 
 Rome loved the intricate exquisite tragedies of 
 the x^^p^^o<f>oi, the dancers. We glibly call 
 them, allow literary people to call them, the 
 decadent successors of the drama. They may, 
 we can believe now, have awoken passionate 
 ecstasies of emotion, beyond our dreams; but 
 they could not be handed down. These "choreo- 
 drames" have perished. So we comfortably fall 
 in with the assumption of those who practise 
 literature, that drama, that queer and monstrous 
 birth, is the God of the theatre. Literary people 
 are very kind to each other; and all-powerful 
 over civilisation. Through them come our his- 
 tory, facts, ideas, and arguments; and so our 
 valuations. We see all things through their 
 
ORIGINS OF DRAMA 41 
 
 mists. The feet of the dancers throb "No !", their 
 heads jerk argument and dialectic to us; we do 
 not heed. We have read of Talfourd, and he will 
 outlive Taglioni. The other arts present them- 
 selves naked, to be accepted as they are. Only 
 literature continually weaves laurels, and is for 
 ever crowning herself. 
 
 But the arts had always an enemy, especially 
 the arts of the theatre. The plays we know of 
 and the dancing we ignore were equally threat- 
 ened by religion, who brought with her the blind 
 forces of asceticism and morality. Any emo- 
 tional and absorbing view of the universe that 
 throws the value of life over into the next world, 
 naturally regards things of this world as means 
 rather than ends. And so it always tends to com- 
 bine with and use that deep instinct in human 
 nature, the instinct to treat all things as means, 
 which is called Puritanism. For eighteen hun- 
 dred years, religion, when it has been strong 
 enough, has persecuted or starved the arts. At 
 times, when it has grown shallow, it has allowed 
 a thin subservient art to flourish beneath it; an 
 art that, ostensibly educating men to be in some 
 way useful, for this life or the next, couldn't help 
 treating them, for a stolen moment, as ends. 
 Such, perhaps, was the pictorial art of the Mid- 
 dle Ages in Italy. But in general the arts have 
 
42 JOHN WEBSTER 
 
 been kept pretty well under, especially the arts 
 of the theatre, creeping slowly out when religion 
 has slept, as in the eighteenth century, or some- 
 times liberated by such splendid bursts of irre- 
 ligion as produced the Elizabethan drama in 
 England. 
 
 The early fathers of the Church embodied the 
 spirit of religion, knew the Will of God, as 
 clearly in this as in most matters. It is amusing 
 to see that Arius alone went so far as pleading 
 for even a Christian theatre. Here, too, he was 
 a lonely light. All the orthodox makers of Chris- 
 tianity were venomous against spectacula. Like 
 children saving up for one great treat. Chris- 
 tians were consoled by Tertullian for the loss of 
 theatres in this world, by the promise of the 
 future spectacle of the exquisite and eternal suf- 
 fering or richly comic writhing of play-actors 
 and dramatists. The forces of evil triumphed. 
 And the theatre was lost more swiftly and com- 
 pletely than the rest of civilisation, when the 
 double night of barbarism and Christianity set- 
 tled down over Europe. 
 
 The long, long rebirth of the Theatre was a 
 process of roughly the same kind in nearly all 
 European countries. But at present I am chiefly 
 concerned with England. For this country the 
 forces that led to the reappearance of theatrical 
 
ORIGINS OF DRAMA 4S 
 
 art and the drama are generally divided into four 
 groups. There were the various travelling min- 
 strels and entertainers; the folk- festivals and 
 folk-plays; the religious drama; and the influ- 
 ence of the classics. The relative importance of 
 some of the earlier fountains of the English 
 drama has been mistaken, through false psychol- 
 ogy. Great weight is always laid on the various 
 popular festivals and games, and the unconscious 
 relics of old religions. They are said to be ex- 
 amples of the beginning of mimetic art. If peo- 
 ple find a participant in a May-festival taking 
 the name of "The Queen," or a member of a 
 dance assuming a personality with the name of 
 "Ginger-breeches," they stretch delighted fin- 
 gers, crying, "The origins of drama!" It is an 
 error. It is not true that "the practice which 
 lies at the root of dramatic art and of the pleas- 
 ure to be gained from it" is "that of pretending 
 to be someone or something else." ^ That is 
 merely what lies at the root of being an actor; 
 and only one of the things even there, as anyone 
 who has known amateur actors can testify. As 
 such, it is but one of the human instincts which, 
 as it happens, enable us to satisfy our love for 
 seeing drama. It has no more to do with "the 
 pleasure to be gained from dramatic art" than 
 
 * C. H. E. L., vol. v., p. 28. 
 
44 JOHN WEBSTER 
 
 the desire for fame which made Keats write, or 
 the desire for expression which made Wagner 
 compose, have to do with poetry or music. They 
 are conditions; at the most, indispensable condi- 
 tions. The point of an art is in the state of mind 
 of the recipient. 
 
 "The poet sings because he must; 
 We read because we will." 
 
 Certain pleasant and valuable states of our minds 
 when we see it, are what distinguishes dramatic 
 art. Only such causes as produced them, or 
 earlier forms of them, are directly relevant to a 
 history of the drama or the theatre. Folk-games 
 and festivals, and even folk-drama, have, there- 
 fore, it seems to me, nearly no relevance to the 
 history of the English drama. 
 
 What is much more important is, of course, 
 the religious drama. Religion, incessantly and 
 half-consciously hostile to the arts, has inces- 
 santly and half-consciously fostered them. 
 Every activity of the mind of man is both end 
 and means; and it is as impossible for religion 
 to confine art to be useful, as it is for the pure 
 "hedonist" to make it merely an end. When the 
 first moralist discovered that by putting his ad- 
 vice into a rhymed couplet he interested and im- 
 pressed the people more, he opened the flood- 
 
ORIGINS OF DRAMA 45 
 
 gates. There soon came along somebody who 
 thought more of the jingle than of the morality. 
 The moralist was powerless to prevent him. 
 Thence follow Martial, Villon, English folk- 
 songs, the Earl of Rochester's play, Baudelaire, 
 and all the abominations of the holy. As the 
 earliest Christian artist sought, in illustrating 
 some incident from Christ's life, to enrich Truth 
 with Beauty, the ghostly, unborn fingers of the 
 Breughels and Felicien Rops guided his brush. 
 
 So while Christianity was busily disinfecting ""7 
 the front hall, the most dreadful smells were j 
 starting again in the scullery. As early as the i 
 fourth century, before she was yet able to tri- 
 umph completely in the defeat of the pagan 
 theatre, the Church had begun to show forth part 
 of the greatest drama in her universe, by repre- 
 sentation, and with all the pomp and wonder of 
 the highest dramatic art. Those who admit the 
 existence of other varieties of theatrical art be- 
 sides the entirely realistic, must recognise that 
 the state of mind of the spectator of the INIass is 
 strongly aesthetic. Other elements enter, but they 
 combine, not clash, wdth this. The fact the spec- 
 tator thinks that what is being represented is 
 true does not make the whole thing undramatic. 
 It becomes a variety of drama, as portrait-paint- 
 ing is a variety of pictorial art, but with less dis- 
 
46 JOHN WEBSTER 
 
 cordant ends than the portraitist must try to 
 serve. That the importance of the Mass is quite 
 other than aesthetic is irrelevant. Considered in 
 the hght of the states of mind of the spectators 
 of that time, the Mass must have been great 
 drama as surely as Giotto's pictures of the life 
 of Christ were great pictorial art. 
 
 Other services and ceremonies of the Church 
 followed in admitting more or less of drama. 
 The history of them, the Quem quceritis trope and 
 the rest, had been worked out and often related. 
 The progress from few to many occasions for 
 gratifying the theatrical instinct in men was in- 
 evitable. More elaborate as well as more numer- 
 ous, as the centuries went on, grew the liturgical 
 dramas. They soon began to be transported out- 
 side the churches; finally to be played by lay- 
 men. More and more scenes from the Bible and 
 from legend were dramatised and performed. 
 They became definitely amusing and interesting 
 for the people, quite apart from the lessons 
 they might teach. Rather too much stress has 
 been laid, naturally, on the great cycles, of Ches- 
 ter, York, Coventry, and elsewhere, that have 
 survived. The accident of their existence must 
 not make us forget that, in church and out, espe- 
 cially l out, there were innumerable miracle and 
 mystery plays continually being played through 
 
ORIGINS OF DRAMA 47 
 
 England in the two or three hundred years be- 
 fore Elizabeth. Every little town and village 
 seems to have had them. They were the ordinary 
 food of the theatrical instincts of the people. We 
 cannot understand them now — what there is left. 
 They are far from our ideas of drama, and by 
 our standards they fail. We can see that some 
 of the episodes were funny, that others had 
 pathetic or tragic value, or a queer vitality of 
 characterisation. But the whole seems incoher- 
 ent, disjointed, and "inartistic." Careful writers 
 go through them, picking out bits of "realistic 
 humour" in one place, and "true literary feeling" 
 in another. It is meaningless; a prattling rela- 
 tion of which parts of these plays appeal to a 
 twentieth century professor. What did those 
 curious mediasvals feel when they were watching 
 them? We cannot tell. They may have had as 
 profound and passionate emotions as a play of 
 Ibsen's stirs in us. But as we do not know we 
 cannot affirm that this mediaeval drama was good 
 or bad; any more than we can for the Greek 
 drama. Which of the two, for instance, was the 
 greater? It is like a deaf mute having to judge 
 whether Strauss or Mozart is the greater opera- 
 maker. Judging from the librettos, and from 
 watching conductors, he might guess that Strauss 
 was more interesting, Mozart more melodious. 
 
48 JOHN WEBSTER 
 
 . . . He could play with inferences. ... So 
 (whatever may be claimed by Greek scholars) 
 must we confess almost complete ignorance about 
 the medieval drama. Some things can be said. 
 It was certainly narrow; and it cannot have had 
 those qualities of concentration and "dramatic 
 unity," that are necessary for great dramatic art 
 as we are used to know it. But I think there 
 may have been, to the contemporary, more con- 
 nection and significance in many of these series 
 of plays than the modern will allow. Or rather, 
 the modern sometimes will admit it intellectually, 
 but he does not realise it emotionally. I can 
 conceive the mediaeval mind (the exceptional 
 mediaeval mind, I admit, for the ordinary childish 
 one must have viewed scene after scene with that 
 transient delight, on a background of reverence, 
 with which schoolboys read Henry the Fourth — 
 they find bits very interesting, and they know it's 
 all for their education) tasting in each episode 
 both the episode itself and the whole, in such a 
 way that, finally, that whole loomed out pecu- 
 liarly solid, majestic, and impressive. The mind 
 would, from its ordinary bent of religious and 
 moral thought, be prepared to receive the play 
 (or cycle) in just this way; and the whole thing 
 would fall into these predestined mental channels 
 with immense accumulating force and power. 
 
ORIGINS OF DRAMA 49 
 
 Just as the Agamemnon was meant for, had its 
 significance for, a mind naturally thinking in 
 terms of v(3pL<s and art]; so, perhaps, a mediaeval 
 series of plays could only find their value in a 
 mind thinking naturally and immediately in 
 terms of the whole Bihlical story, theologically 
 interpreted. To the Greek mind the rugs laid 
 down for Agamemnon trailed clouds of horror; 
 to the mediaeval the incident of Cain and Abel 
 may have suggested straightly and sincerely, in 
 a way we could never feel it, the entire ancestry 
 of Christ, or the meaning of a later greater sacri- 
 fice, and may have illuminated and caught light 
 from the whole tremendous process of the work- 
 ing out of the Will of God. I do not know if 
 the mediaeval cycles consciously tried to produce 
 an effect of this kind, or if they ever succeeded, 
 enough to make them worthy, in their narrow 
 kind, to stand by the great dramatic products of 
 other styles and other ages. I only suggest that, 
 aesthetically, they may have been of this nature. 
 It is a method, this subordinating the parts to the 
 whole, in such a way that the parts have no neces- 
 sary connection with each other except through 
 the whole, that is strange to us who are used to 
 "plots" that centre about one incident or situa- 
 tion, or one or two characters. In it Time or Fate 
 is the protagonist. It might have, but never did. 
 
50 JOHN WEBSTER 
 
 come off in those dreary chronicle-plays, that 
 increase the desolation of the early Elizabethan 
 drama. It is a method that has been used in 
 later days with greater success. Wagner in The 
 jRing gets something of this effect. And Hardy 
 in TJie Dy7iasts and Schnitzler in Der Junge 
 Medardus have used these apparently discon- 
 nected, episodic scenes, with or without commen- 
 tary, for a resultant whole as different from 
 them as a face is from its parts, nose, eyebrows, 
 ears and the rest. They show you a street-scene, 
 some friends, two lovers — all irrelevant — and 
 you know Vienna of 1809. Or they pick out, 
 perhaps, and light up, a few disconnected objects 
 on the stream of time, and you are suddenly, ter- 
 ribly aware of the immense black unreturning 
 flood, sliding irrevocably between darknesses. 
 
 Such a method, however, if it existed in 
 mediaeval times, did not influence the Elizabethan 
 drama. The disconnected narrative form was in- 
 deed an Elizabethan inheritance from mediaeval 
 religious drama; but merely as narrative. The 
 narrative was transferred from sacred subjects 
 to historical; the line is pretty clear. The chron- 
 icle-plays, indeed, appear to be artistically a 
 retrogression. In incidents and in the whole 
 they are more pointless. The loose narrative 
 style, the limber and many- jointed acts, and the 
 
ORIGINS OF DRAMA 51 
 
 habit of bringing everything on the stage, lasted 
 in the plays of the great period — the beginning 
 of the seventeenth century. Besides this, the mir- 
 acle and mystery plays gave little to the Eliza- 
 bethan drama. They handed on the possibility 
 of tragedy and comedy; but that gift was not 
 needed. They bequeathed, too, a certain rather 
 admirable laxity and vagueness with regard to 
 locality in drama; and a tiresome, confusing 
 tendency to make plays illustrate a moral, a ten- 
 dency which fitted in only too well with the 
 theory of Elizabethan times; less, fortunately, 
 with its practice. 
 
 These miracles and mysteries in their various 
 forms lasted, in country parts at least, to over- 
 lap with the Elizabethan drama. But there was 
 another form of the religious play which actually 
 formed the chief link with the later style, the 
 morality. It was a late growth, and it rather 
 superseded the miracles and mysteries. It was 
 aided, though not originated, by the revival of 
 learning and moral fei^our that followed the 
 Renascence and accompanied the Reformation; 
 and, coming at this time, it soon widened from 
 merely religious ideas to all kinds of secular in- 
 tellectual notions. It is distinctly of the age 
 of Protestantism, and so we can understand it, 
 better at least than its predecessors, in the same 
 
52 JOHN WEBSTER 
 
 way that we can understand Erasmus. It deals 
 less with God and more with man and the ab- 
 stractions that were thought to surround his life. 
 By such strange ways the arts came home. 
 JNIoralities and moral interludes, in their turn, 
 could have produced ( and did produce in Every- 
 man at least) great drama in their kind. But 
 again, it was a narrow kind. Had that tide 
 flowed on unchecked, we might now look back 
 on an immense English Drama of types and 
 personifications, a noble utterance, in this nar- 
 row sort, of all the human desires and dreams 
 and interpretations of life for centuries. The 
 crown and glory of the English theatre would 
 have been Milton — Comus, even now, is, in dis- 
 guise, the most noble example of morality. We 
 might have achieved the most solemn and noble 
 drama of the world — a nobility astonishingly dif- 
 ferent from the glory we have achieved, its direct 
 opposite. For the transformation of the moral- 
 ity into the Elizabethan play was a complete re- 
 versal of direction. The whole point of the 
 former is that it deals with the general; you 
 find all your experience drawn together and illu- 
 minated; you are pervaded, rather than shaken, 
 with the emotion of the philosopher who sees the 
 type through the individual. Love beneath the 
 lover. The latter gives you the particular; some 
 
ORIGINS OF DRAMA 53 
 
 definite person or circumstance so poignantly 
 that you feel it; the reality for those vaporous 
 abstractions, not Love but William in love, not 
 Death but some fool, rather untidily, dying. The 
 one shows you Everyman, the other Hamlet. 
 Each way is good; but to go from one to the 
 other, is as if English art twenty-five years ago 
 had suddenly swung from Watts to Whistler. 
 
 Those who are fond of comparing epochs in 
 history with stages in the life of a man will be 
 pleased to liken the medieval miracles and mys- 
 teries to the narratives that delight children, the 
 period of the moralities to that invariable love 
 of youth for generalities and proverbial wisdom 
 — for Love, Death, Fate, Youth, and all the 
 wonderful heart-lifting abstractions — and in the 
 Elizabethan's climb to that chief abode of art, 
 the heart of the individual, they will find the 
 middle-aged turning, with the strength as well 
 as the bitterness of agnosticism, to all that on^^ 
 can be certain of, or, after a bit, interested in, / 
 men, women, places, each as a "special case.' 
 But if the moralities are taken on their own 
 merits and not as a step in a process, it is doubt- 
 ful whether they are, artistically, an advance on 
 miracles and mysteries. Dodsley's point, that 
 they were a better kind, as giving the author 
 greater freedom, enabling him to invent his plots, 
 
54 JOHN WEBSTER 
 
 has been often repeated. There is not much in 
 it. The Greeks and most of the Elizabethans did 
 not, in that sense, "invent" their plots. In the 
 Christian stories and legends the greatest drama- 
 tist could have found enough to last him a life- 
 time. Any old story does for the framework of 
 a play. The moralities, in fact, in putting the 
 dramatist to the trouble of inventing a "plot," 
 rather tended to divert his attention from more 
 important things. In other ways, however, they 
 did widen the ground for the dramatist; and in 
 making plaj^s more wholes and less narratives, 
 and insisting on dramatic unity, they prepared 
 very efficiently for the Elizabethan kinds of 
 drama. It might, indeed, have been better if their 
 legacy of dramatic unity had been more strictly 
 observed. Their other characteristic, of thinking 
 in types and abstractions, instead of individuals, 
 had a longer influence, of no very healthy kind, 
 than is at first obvious. Dr. Faustus is only 
 Everyman, or at least Every-philosopher, with 
 a name and a university degree. And there was 
 also a moralising effect; which is not quite the 
 same thing. An art which proceeds by personi- 
 fications of abstract ideas need not moralise, 
 though in this instance it nearly always did. A 
 modern morality in which the characters were 
 Evolution, The-Survival-of-the-Fittest (his 
 
ORIGINS OF DRAMA 55 
 
 comic servant), Man, and the various Instincts, 
 might be very impressive without conveying any 
 moral at all. The Elizabethan drama, however, 
 started with the burden of this idea among 
 others, that a play rather ought to specify a 
 moral generalisation. It took some time to shake 
 it off. 
 
 The third more or less dramatic activity 
 through the Middle Ages was provided by the 
 minstrels and strolling entertainers of various 
 kinds. The ancesters of these were on the one 
 hand the actors of Rome, the mimi, who, when 
 the theatres ceased, took to wandering about and 
 giving entertainments, and on the other the more 
 reputable and probably less dramatic Teutonic 
 scop. These minstrels were a great feature of 
 the whole medieval period, but their importance 
 in the history of the theatre has always been 
 under-estimated. There are two reasons, I think. 
 One is that their performances have left very lit- 
 tle record. The history of religious drama can 
 be traced fairly fully. Minstrels of all kinds 
 may have been giving unceasing dramatic enter- 
 tainments throughout Europe during the same 
 centuries. We have nothing to say about it. 
 There are no traces to investigate, no written 
 text of the performances to comment on. So, as 
 we continually hear of the religious perform- 
 
56 JOHN WEBSTER 
 
 ances and never of these others, we insensibly 
 grow to attach great importance to the former 
 and to omit the latter altogether in our view. 
 The second reason lies in the error in psychology 
 I have discussed. It is supposed that, while any 
 band of rustics dressing up is relevant to the his- 
 tory of drama, no entertainment given by min- 
 strels is, unless it is full-blown realistic acting. 
 I think that careful consideration of the imagined 
 states of mind of a mediaeval, or indeed of a 
 modern, audience, will show that the theatrical 
 emotion begins far before that. Even a single 
 minstrel reciting a tragic story seems to me 
 nearer to evoking it than many apparently more 
 "mimetic" activities. And directly he introduces 
 any representation or imitation — as reciters 
 always tend to do — drama is, in embryo, there. 
 I think it is certain that a single performer can 
 produce all the effects of drama, by represent- 
 ing, conventionally, several characters in turn. 
 Mile. Yvette Guilbert does it. You get from 
 her the illusion of seeing, with extraordinary 
 insight and vividness, first the prisoner of 
 Nantes, and then the gaoler's daughter, quite as 
 much as you would in an opera. The thing can 
 go further. I myself have seen a mere amateur 
 represent at one time and in his one person two 
 lame men, each lame in a different way, walking 
 
ORIGINS OF DRAMA 57 
 
 arm-in-arm, with almost complete realism. And 
 when it comes to dialogues and estrifs between 
 two or more performers, it seems to me absurd 
 pedantry, a judging by forms instead of realitie^_ 
 to deny the presence of drama. 
 
 In any case, the miifii went into the darkness, 
 at the end of Rome, performing plays; and the 
 same class reappears, performing plays, as soon 
 as we can discover anything about them, cen- 
 turies later. The influence of the farces these 
 wanderers were playing towards the end of the 
 middle ages, on early English comedy, is more 
 or less recognised. I think it is very probable 
 they had a great influence also on tragedy and 
 on drama as a whole. Some of them, it is known, 
 used to perform puppet-plays wherever they 
 went. The importance of these in keeping drama 
 and the taste for tragedy and comedy alive in 
 the hearts of the people is immense. These 
 strolling professional entertainers took their part 
 also in other kinds of dramatic performances. 
 We find them helping in folk-plays and festi- 
 vals; and when the religious plays were secular- 
 ised, they often appear as aiding the amateurs. 
 Indeed, the "interlude," the favourite dramatic 
 form which develops out of the secularised relig- 
 ious plays, and which led straight to the Eliza- 
 bethan drama proper, fell largely into the hands 
 
58 JOHN WEBSTER 
 
 of the "minstrels." About that time they were 
 reinforced, and rivalled, by the various local com- 
 panies of actors v^ho began touring in a semi- 
 professional way. They were also strengthened 
 during the fifteenth and sixteenth centuries by 
 being enrolled in the service of various great 
 lords. Under both popular and aristocratic cir- 
 cumstances these professionals, after severe com- 
 petition with amateurs during the first part of 
 the sixteenth century, settled, some of them, into 
 theatres, and became the actors of the Eliza- 
 bethan drama. Their importance in this light is 
 obviously very great. But their true position can 
 be guessed by inspecting Mr. Chambers' appen- 
 dices of medieval plays and Mr. Tucker Mur- 
 ray's more recent researches. It was they that 
 were responsible for continual dramatic perform- 
 ances of every kind throughout England. How 
 good or bad these were vv^e cannot tell. The 
 forces of religion opposed them, with varying 
 vigour at different periods, and probably suc- 
 ceeded in degrading them to a low level. But 
 they must have prepared the mind of the people 
 to exj^ect certain things in tragedy or comedy; 
 and they may account for various aspects of 
 Elizabethan plays that neither the religious nor 
 the classical influence explains. 
 
 By the middle of the sixteenth century, then, 
 
ORIGINS OF DRAMA 59 
 
 the drama was in an inchoate condition. Inter- 
 ludes of all kinds, moral, religious, controversial, 
 and farcical, were being played by all sorts of 
 audiences, besides the rough beginnings of popu- 
 lar tragedy and comedy, and many survivals of 
 the old religious plays. In the sixties the real 
 Elizabethan drama began; and one of the chief 
 influences in working the change was the classical 
 one. It came from above, and from amateurs. 
 It was started, it is noteworthy, by people with 
 a fixed, conscious, solemn, artistic aim. They 
 wanted to have tragedies in the real classical way ; 
 so they imitated, queerly enough, Seneca! Eng- 
 lish Hterature has always been built on a rever- 
 ent misunderstanding of the classics. Anyhow, 
 anyone is good enough to be a god. The worst 
 art has always been great enough to inspire the 
 best. The iron laws of heredity do not affect 
 literature; and Seneca may father Shakespeare 
 as Macpherson fathered the Romantic Move- 
 ment. 
 
 The dates of the Senecan movement in Italy, 
 France, and England have been elaborately 
 worked out. They do not concern us now. The 
 influence of Seneca, and, vaguely, what was 
 thought to be the classical tradition, in accord- 
 ance with the misunderstood laws of Aristotle, 
 came primarily by two streams, through Italy 
 
60 JOHN WEBSTER 
 
 and France. Tancred and Gismunda was influ- 
 enced by the Italian Senecans; Kyd translated 
 Gamier. Italy, of course, the romantic home 
 of all beauty and art, had the most influence. But 
 culture came from France. The English began 
 translating Seneca for themselves in the sixties 
 and seventies. As far as can be seen, the posi- 
 tion in the eighties, when Marlowe and Kyd were 
 about to fling English tragedy as we know it 
 shouting into the world, was that the popular 
 stage was scarcely touched at all by this classical, 
 Senecan movement; the children's companies and 
 ordinary court plays were only partly and patch- 
 ily affected; but private performances in the 
 Inner Temple and Gray's Inn had proudly and 
 completely adopted the Senecan (or, generally, 
 classical) style. As these were often given be- 
 fore the Queen, they had great influence in 
 spreading the impression that this type of trag- 
 edy was the highest, the only type intellectual 
 and cultivated people could aspire to. The Sene- 
 can boom did not leave much directly to Eliza- 
 bethan drama; far less than is generally made 
 out. It left perhaps a ghost tradition, the much- 
 advertised and over-valued "revenge motive," 
 and the tendency to division into five acts. But 
 indirectly it had value in tightening up the 
 drama, pulling the scattered scenes which appeal 
 
ORIGINS OF DRAMA 61 
 
 to the English, a httle, but not too much, into 
 one play. And it was of vast use as an ideal. 
 It enabled the dramatists to write for their audi- 
 ences but above them. It set the audiences an 
 aesthetic standard, shook them into artistic moral- 
 ity. Left to itself, this movement would have, 
 and did, become academic, cold, dead. But 
 Fulke Greville, Alexander, even Ben Jonson, 
 did not get the full benefit of it. The best of it, 
 and the best of the popular stage, were torn out, 
 combined, and revitalised by Kyd and Marlowe. 
 Towards that the times were ripening. The 
 drama was getting a standing, the first important 
 step. It was at once popular and fashionable. 
 And, though a few Puritan fanatics had started 
 a protest, the main mass of the people were 
 against them. That gradual depletion of the 
 theatre-audiences which took place during the 
 next century, when bourgeois democracy slowly 
 became one with Puritanism, had not com- 
 menced. The establishment of fixed theatres in 
 London must have raised the level of the per- 
 formances; and, the second important step, it 
 was educating and preparing an audience. For 
 an audience must be trained and trained together, 
 as much as a troupe of actors. It is equally one 
 of the conditions of great drama. 
 
Chapter III 
 
 THE ELIZABETHAN DRAMA 
 
 There are many ways of considering a subject 
 like the Elizabethan drama. You can take the 
 plays by authors. Naturally, it is one of the best 
 ways ; and it is the only way that was employed 
 up to quite recently. To use that method alone 
 leads to queer blindnesses. And it is apt to end 
 in the "our Shakespeare" business, an easy and 
 unprofitable way of taking art. 
 
 Then there is division by subjects, the method 
 of Professor Schelling and of Polonius. This 
 counteracts the evils of the first way; but it is 
 often rather unmeaning: Measure for Measure 
 gets grouped with the "Romantic Comedies." 
 That is to say, the fault is in the unreality of the 
 classes. They should rather be grouped by taste. 
 An arrangement under purely fanciful names 
 would be more practical. Love's Labour Lost 
 would go with Lyly under "Court Butterfly"; 
 Measure for Measure might jostle The Fawn or 
 Hamlet in the "Brass-on-Tongue" sub-division 
 of the "Leaves-a-Taste-in-the-Mouth" group. 
 
THE ELIZABETHAN DRAMA 63 
 
 And there is the reader's way, liamb's way, 
 of just picking out the best plays. It has a lot 
 to be said for it. 
 
 All three methods, and others, have their com- 
 plemental merits. But I think the most useful 
 way of surveying material like this is by a com- 
 bination, in the following way. One should 
 divide the plays, roughly chronologically, accord- 
 ing to their style or taste, the general Stimmung 
 of them, with a certain reference to authorship, 
 and distinct emphasis on the merits and possibili- 
 ties of the various styles. For though, of course, 
 when you stop to consider any particular part, 
 these questions of influence, "schools," styles, 
 periods, and the rest, immediately sink into their 
 proper subordination, yet, for a rapid survey, 
 they do correspond to certain realities. It is 
 important to know that a writer was aiming at a 
 certain atmosphere, or influenced by it. And 
 some of these atmospheres, and these aims, are 
 much healthier for art than others. At any rate, 
 I think that to explain what Webster's plays 
 really are, it is necessary to show where they fit 
 in with the rest of the Elizabethan drama. And 
 as I do not know of any survey of this drama 
 that seems to show the main outlines right, espe- 
 cially with regard to comparative goodness — the 
 scientific literary historian makes every play 
 
64 JOHN WEBSTER 
 
 equally dull, the Swinburnian critic makes every 
 author equally supreme — I shall try to give, very 
 briefly, my own views. 
 
 Soon after Lyly began to breathe into comedy 
 (with which I am not concerned) a movement 
 that was near to being life, and a prettiness that 
 was still nearer beauty, Kyd and Marlowe blew 
 life, strength, and everything else into tragedy. 
 To say that they grafted the energy of popular 
 tragedy on the form of classical, would be to 
 wrong by a soft metaphor their bloody and vital 
 violence. It was rather as if a man should dash 
 two dead babies together into one strident and 
 living being. Kyd, of course, does not really 
 stand by Marlowe. But he seems further below 
 him than is fair, because Marlowe's genius was 
 more literary, and so lives longer. Both brought 
 light and life to tragedy. Kyd filled Seneca's 
 veins with English blood. He gave his audience 
 living people, strong emotions, vendetta, murder, 
 pain, real lines of verse, and, stiffly enough, the 
 stateliness of art. He thrilled a torch in the 
 gloom of the English theatre. Marlowe threw 
 open a thousand doors, and let in the sun. He 
 did it, in the prologue to Tamhurlaine, with the 
 superb insolence and lovely brutality of youth. 
 His love of the body, his passion for the world 
 of colour and stuff, his glorious atheism, "giant- 
 
THE ELIZABETHAN DRAMA 65 
 
 ism 'gainst Heaven," were trumpets in that 
 morning. The blood still sings to them. Mar- 
 lowe is less representative, stands clearer of his 
 period, than almost any Elizabethan. He was 
 of no school, had no followers. Others, Shake- 
 speare, for instance, caught something of his 
 trick of blank verse, or tried a play or two in 
 his manner. But there was no body of drama 
 that partook of the atmosphere of ferocious, 
 youthful, passionate tragedy that distinguishes 
 Marlowe's work. He stands rather, in his joy 
 of the world, and irreligion, as the herald of the 
 whole age, and of that short song of passion it 
 could utter before the beginning of the night. 
 His loneliness is explicable. It was not only 
 that no contemporary was old and great enough 
 to take all he had to give. But his dramatic 
 method was unique. He was not a dramatist in 
 the way the others were. He was — in this some- 
 thing like the young Shakespeare, but far more 
 so — a lyric writer using drama. "Plot" does not 
 matter to him. Each scene he works up into an 
 intense splendid lyric. They are of different 
 kinds, but put together they have unity. The 
 whole is a lyric drama. No one else, except, con- 
 ceivably, Webster, in a slight degree, used this 
 artistic method. Marlowe was an extreme poin- 
 tilliste. He produced his whole effect by very 
 
66 JOHN WEBSTER 
 
 large blobs of pure colour, laid on side by side. 
 The rest were ordinary semi-impressionists, with 
 a tale to tell. Only Webster more than rarely 
 achieved expressionism. 
 
 One other gift Kyd and Marlowe, especially 
 JMarlowe, gave their contemporaries; blank 
 verse. Before them was the Stone Age; they 
 gave the poet a new weapon of steel. Marlowe 
 was drunk on decasyllabics, the lilt and clang and 
 rhetoric of them. How he must have shouted, 
 writing each line of Tamhurlaine! It all fits in 
 with the rest of this outburst of true great trag- 
 edy in the eighties. But it was only an outburst 
 of youth ; and the sentimentality and tediousness 
 of youth had to be gone through before the best 
 times could be won. The rest of the history of 
 the drama during this century is mainly con- 
 cerned with the histories and chronicles. Some- 
 thing — it may have been the Spanish Armada — 
 made the audiences demand this dreary kind of 
 play. Their other cry (I have only space to dis- 
 cuss the best audiences and plays) seems to have 
 been for a slight kind of romantic comedy. They 
 swallowed everything, of course, as at all periods 
 of this eighty years. But these two types of 
 play, were, perhaps, most prominent. 
 
 Critics have always idiotically thought it their 
 duty to praise these histories; partly because 
 
THE ELIZABETHAN DRAMA 67 
 
 Shakespeare, in obedience to popular demand, 
 wrote some; partly because they are supposed 
 to exemplify the patriotism of the Elizabethans, 
 and we are supposed to enjoy that patriotism. 
 These chronicle-plays fit in, it is not very clear 
 how, with Drake, Hawkins, and the rest of the 
 "island story." And those numerous literary or 
 dramatic critics who do not care for literature or 
 the drama, nod their sentimental approbation. 
 It sounds too fantastic for truth, but it is true, 
 that the ultimate defence of Elizabethan drama 
 offered by many writers on it, is that it holds up 
 so faithful a glass to the "bustling, many-sided 
 life of that wonderful time." Such wretched 
 antiquaries beam mild approval on these new 
 proofs of the Elizabethan's interest in his coun- 
 try's history. 
 
 It must be clearly decided that these histories 
 were a transient, dreary, childish kind. They 
 preserved the worst features of Elizabethan 
 drama in their worst form ; the shapelessness, the 
 puerility, the obvious moralising, the succession 
 of scenes that only told a narrative, the entire 
 absence of dramatic unity, the mixture of farce 
 and tragedy that did not come off. I do not 
 mean (for the moment) to say that the Eliza- 
 bethan type of play was bad, as such; only that 
 when done in this form it was silly and without 
 
68 JOHN WEBSTER 
 
 value. One or two tragedies that were written 
 in the form of histories are some good; Richard 
 II and Edward II, And, of com'se, in his worst 
 efforts Shakespeare always leaves touches of 
 imagination and distinction. But as a whole 
 these histories are utterly worthless. 
 
 Something similar is the case with the romantic 
 comedies. Neither in themselves, nor as a sign 
 of the taste of the times, have they much value. 
 Occasionally they achieve a sort of prettiness, the 
 charm of a stage-spring or an Academy allegory 
 of youth. And Shakespeare threw a pink magic 
 over them. But it should be left to girls' schools 
 to think that the comedies he obligingly tossed 
 off exist in the same universe with his later 
 tragedies. The whole stuff of this kind of play 
 — disguises, sentimentality, girls in boys' clothes, 
 southern romance — was very thin. It might, 
 perhaps, under different circumstances, have 
 been worked up into exquisite, light, half -passion- 
 ate comedy of a limited kind. It did not achieve 
 even this success. 
 
 There are one or two isolated good plays of 
 indefinable genus, like A Midsummer Niglifs 
 Dream. But on the whole this period of silliness 
 or undistinguished prettiness between the great 
 years of Marlowe (c. 1588) and the wonderful, 
 sultry flower-time of the next century, is only 
 
THE ELIZABETHAN DRAMA 69 
 
 redeemed by one kind of drama that was seri- 
 ously trying to move serious artistic emotions. 
 It is a kind that is despised by the refinement of 
 modern criticism, condemned by the word 
 "crude"; what is called "domestic tragedy." 
 These indigenous plays, descendants probably of 
 unknown myriads of popular tragedies in Eng- 
 land, were nearly always dramatisations of re- 
 cent occurrences. Some are bad, and all are as 
 "crude" as life. But they kept people in touch 
 with realities, with the brutality of blood and 
 death. The theatre might so easily have gone 
 irrevocably soft during these years. They kept 
 it fit for the tragedy that was to come ; and they 
 profoundly influenced that tragedy for the eighty 
 years of "Elizabethan drama." But it was at 
 this time that they were especially common. The 
 only long study of the subject ^ contains a list of 
 the plays of this nature. There are twenty-four 
 known; fourteen of them occur in the period 
 1592-1603, two earlier, eight later. It is note- 
 worthy that of the three best we know, one, 
 Arden of Fever sham, comes at least at the begin- 
 ning of the period, almost in Marlowe's time; 
 the second, A Woman Killed with Kindness 
 
 * Das hilrgerliclie Trauersjnel in England. Singer. The list 
 counts Arden of Feversham as 1592. It is probably earlier, 1586 
 or so. 
 
70 JOHN WEBSTER 
 
 (Hey wood's best play), comes right at the end, 
 in the golden years of the next century, and the 
 third, A Yorkshire Tragedy, is generally dated 
 as right in the middle of that great age, in 1605. 
 
 For there was a period — 1600-1610 are the 
 rough inside limits — that stood out an infinity 
 above the rest. Nearly all the good stuff of 
 Elizabethan drama was in it or of it. Except 
 in comedy, there are only the lonely spring of 
 Marlowe and the Indian summer of Ford out- 
 side it. And it is not only that it was Shake- 
 speare's great time. That is partly both cause 
 and effect, and our great good fortune. 
 
 The whole age, in drama and beyond, was alive 
 with passion and the serious stuff of art. Nor 
 was it only that so much of great merit was pro- 
 duced in this short time. Nearly all the work of 
 the period shared, apart from its goodness, in a 
 special atmosphere. It is extremely important 
 to recognise the absolute distinctness and su- 
 preme greatness of this period, its sudden ap- 
 pearance and its swift and complete end. There 
 is only space here to hint at its characteristic 
 features. It was heralded (poetry is generally 
 a few years ahead of drama) by Shakespeare's 
 sonnets, and the poems of Donne — who, in spite 
 of Ben Jonson, did not write all his best things 
 before 1598. Poets, and men in general, had 
 
THE ELIZABETHAN DRAMA 71 
 
 reached a surfeit of beauty. The Renaissance 
 joy in loveliness, the romantic youthfulness of 
 the age, the wave of cheerful patriotism, all 
 passed at the same time. Boyhood passed. Im- 
 agination at this time suddenly woke to life. Its 
 flights were to the strangest corners and the 
 pitchiest barathrum of the deep. Intellect was 
 pressed into the service of the emotions, and the 
 emotions were beaten into fantastic figures by 
 the intellect. The nature of man became sud- 
 denly complex, and grew bitter at its own com- 
 plexity. The lust of fame and the desire for 
 immortality were racked by a perverse hunger 
 for only oblivion; and the consummation of 
 human love was observed to take place within 
 the bright, black walls of a flea. It seemed 
 as though all thought and all the arts at 
 this time became almost incoherent with the 
 strain of an inhuman energy within them, and a 
 Titanic reaching for impossible ends. Poetry 
 strove to adumbrate infinity, or, finding mysti- 
 cism too mild, to take the most secret Kingdom 
 of Heaven by storm. Imagination, seeking 
 arcane mysteries, would startle the soul from its 
 lair by unthinkable paradoxes. Madness was 
 curiously explored, and all the doubtful coasts 
 between delirium and sanity. The exultations of 
 living were re-invigorated by the strength of a 
 
72 JOHN WEBSTER 
 
 passionate pessimism ; for even scepticism in that 
 age was fecund and vigorous, and rejoiced in the 
 whirling gloom it threw over life. The mind, 
 intricately considering its extraordinaiy prison 
 of flesh, pondered long on the exquisite tran- 
 siency of the height of love and the long decom- 
 position that death brings. The most gigantic 
 crimes and vices were noised, and lashed immedi- 
 ately by satire, with the too-furious passion of the 
 flagellant. For Satire flourishes, with Trag- 
 edy, at such times. The draperies of refinement 
 and her smug hierarchy were torn away from the 
 world, and Truth held sway there with his ter- 
 rific court of morbidity, scepticism, despair, and 
 life. The veils of romanticism were stripped 
 away: Tragedy and Farce stood out, for men 
 to shudder or to roar. 
 
 In a time so essentially healthy for all that is 
 fine in man, and especially in his arts, it is no 
 wonder that the best in a great many different 
 styles was being done. But each of these bests 
 has some trace of the spirit of the times. Chap- 
 man, for instance, was doing his finest serious 
 work. Bussy D'Amhois comes near the begin- 
 ning of the period, the two Byron plays later on, 
 The Revenge of Bussy at the end. Chapman is 
 of the time in his intellect, but not in his emo- 
 tions. His devotion to the "Senecal man," and 
 
THE ELIZABETHAN DRAMA 73 
 
 the archaistic austerities of his style, are his 
 alone. He was too moral for the morbidity of 
 the others, and too dispassionate for their gloom. 
 He was not interested in the same feelings. But 
 his mind delighted in the same intricate convolu- 
 tions of thought and half-absurd, serious para- 
 doxes. And occasionally he strikes into those 
 queer horrors that delighted Donne and Mar- 
 ston, and Tourneur and Webster and Shake- 
 speare. He never made a great success of drama, 
 because he thought in a literary and rhetorical 
 rather than dramatic way. He is good reading, 
 but he would not be good seeing. 'There are two 
 ways of displaying character in literary drama, 
 through words and through action. Chapman 
 has only the first ; Webster had something of the 
 second too. Webster revered Chapman, but he 
 was not much influenced by him. Ben Jonson 
 also is at first sight apart from the spirit of this 
 period, although his best work belongs to it. His 
 theories of tragedy prevented him from con- 
 tributing to the Marston-Tourneur- Webster 
 type of play. He would have condemned the 
 atmosphere which is their great virtue as un- 
 classical. They probably did so — we know Web- 
 ster did so — themselves. But he is very relevant, 
 all the same. In the first place that attitude of 
 professionalism in art and respect for the rules 
 
74 JOHN WEBSTER 
 
 which he stood for all his life, was a great factor 
 in raising the dignity of drama and the standard 
 of the dramatists. But Jonson's chief influence 
 and achievement in English drama was in found- 
 ing the Comedy of Humours ; and both this kind 
 of play and his examples fit in with the rest of 
 the time. It is so far from sentimentalism, such 
 a breaking with romantic comedy, this boisterous 
 personification of the "humours" of mankind, 
 with its heartiness and rough strength. It has 
 the life of the time. Jonson brought comedy 
 home to England and to men. The characters 
 in his comedy were not complete men, but they 
 were human caricatures, the right stuff for farce 
 and loud laughter. Their vigour grew amazing 
 under his handling. In result he gave the stage 
 the best comedies of all the age. Their coarse 
 splendour of life was never approached till 
 twenty years or more had passed, and his influ- 
 ence again was strong, in the work of some of 
 his "sons." There, comedy survived the floods of 
 sweetness under which tragedy utterly perished. 
 But if Epicoene and The Alchemist are admi- 
 rably complementary in this Pantheon to 
 Sophonisha and The Duchess of Malfi and 
 Timon of Athens and Macbeth, other works of 
 Jonson are something more. It is probable that 
 the additions to the 1602, The Spanish Tragedy, 
 
THE ELIZABETHAN DRAMA 75 
 
 are Jonson's. If so, he is responsible for some 
 of the finest scenes of imaginative horror in that 
 hterature. These few pages (written in 1600) 
 contain most of the terror and splendour of the 
 next ten years. They set the tune unfalteringly. 
 And Jonson did also what Marston never quite 
 succeeded in doing, he wrote a good comedy 
 which had more of this seventeenth century 
 pungency in it than any tragedy, a comedy that 
 is a real companion to the tragedies of Webster. 
 The mirth of Tourneur is horrible; Languebeau 
 SnufFe poises one sickly between laughter and 
 loathing. Volpone is like one long laugh of 
 Tourneur's, inspired by a tenfold vitality. It is 
 amazing, one of the few complete works of genius 
 of the Elizabethan age. The hot cruelty and 
 vigorous unhealthiness of it! Its very artistic 
 perfection is frightening and exotic. 
 
 But perhaps the main current of strength in 
 the drama during these years, and certainly the 
 most important for this essay, is that which ran 
 through Marston and Tourneur to Webster. 
 Donne was in connection with it, too, from the 
 side of poetry and thought. The relation of 
 Shakespeare with the whole of this period, of 
 which he, then at his greatest, was, to our eyes, 
 the centre, is curious. His half -connections, the 
 way he was influenced and yet transmuted the 
 
76 JOHN WEBSTER 
 
 influences, would require a good deal of space to 
 detail. But in this, his "dark period" — whatever 
 it was, neuralgia, a spiritual crisis, Mary Fitton, 
 or literary fashion, that caused it — he was not 
 unique or eccentric in the kind of his art. His 
 humour was savage, he railed against sex, his 
 tragedies were bloody, his heroes meditated curi- 
 ously on mortality. It was all in the fashion. 
 His gloom was not conspicuous in the general 
 darkness. He had, in Hamlet especially, affini- 
 ties with this ISIarston- Webster group. His ter- 
 rific and morbid studies of madness influenced 
 theirs. 
 
 Marston is one of the most sinister, least un- 
 derstood, figures in Elizabethan literature. More 
 than anybody else, he determined the channels 
 in which the great flood of those ten years was to 
 flow. His life was curious. He started, like so 
 many of them, by writing vivid, violent, crabbed 
 satire. He went on to play-making, which he 
 pursued for eight years with great success. He 
 was much admired and very influential, but he 
 always presented himself to the world with a typ- 
 ical, passionate ungraciousness. At the end of 
 the eight years he renounced the applause that 
 he so liked disliking, and went into the Church. 
 He had a queer lust for oblivion. His tombstone 
 bears Oblivioni Sacrum, It was his personality 
 
THE ELIZABETHAN DRAMA 77 
 
 rather than his powers that was the most stupen- 
 dous thing about him. To us he seems nearly 
 always just not to bring his effects off; but his 
 contemporaries, whatever they thought, could 
 not escape him. 
 
 He started the movement of this period by 
 resuscitating the old blood-and-thunder revenge 
 tragedy. It was precisely what was needed, but 
 he clothed it with his own peculiar temperament 
 of violent and bloody satire. It was this that 
 really attracted the writers of the time. He gave 
 them several plays steeped in it, both comedies 
 and tragedies by the ordinary classifications, 
 really only of one kind. The horror and inhu- 
 man violence of his laughter lit up those years 
 like a vivid flash of lightning. He is responsible 
 for that peculiar macabre taste, like the taste of 
 copper, that is necessary to, if it is not the cause 
 of, their splendour. But he was of his age in its 
 strength as well as in its morbidity. 
 
 "My God's my arm ; my life my heaven, my grave 
 To me all end," 
 
 says Syphax. Chapman could scarcely have 
 equalled the strong nobility of it. 
 
 Marston's chief passion was for truth. He 
 preferred it if it hurt; but he loved it anyhow. 
 It comes out in the snarling speculations and 
 
78 JOHN WEBSTER 
 
 harangues of those satirical malcontents he was 
 so fond of. He bequeathed the type to Tourneur 
 and Webster. For Marston, who was a wit and 
 a scholar and a great poet, was pre-eminently a 
 satirist. It was because he loved truth in that 
 queer, violent way that some men do love, desir- 
 ous to hurt. It fits in with his whole tempera- 
 ment — vivid, snarling, itching, dirty. He loved 
 dirt for truth's sake; also for its own. Filth, 
 horror, and wit were his legacy; it was a splendid 
 one. Some characters too, besides the Malcon- 
 tent, were his offspring. He may have origi- 
 nated the heroine who was wicked or non-moral, 
 fascinating and not a fool. It was a type that 
 was refreshingly and characteristically promi- 
 nent in the great period. Cleopatra, Vittoria, 
 the Insatiate Countess — the womanly heroine 
 fades to a watery mist when they sweep on. 
 Marston is more famous for what he lent than 
 what he had, but what he had is superb. 
 
 Of Tourneur (the dates of whose play, or two 
 plays, are most uncertain) less need be said. 
 Nowadays he is thought better than JNIarston. 
 He is really far his inferior. He does not shock 
 you in the same way by hideously violent con- 
 trasts. He is more level; he is more conscious 
 of his purpose; and it may be ti'ue that none of 
 Marston's plays is as good as his (if he did write 
 
THE ELIZABETHAN DRAMA 79 
 
 The Revenger's Tragedy) . But Marston is the 
 greater genius. Still, Tourneur with his brilliant 
 and feverish morbidity carried on the line. He 
 did not influence Webster so deeply as Marston 
 did. It was natural. He used for the most part 
 external horrors for horror's sake. He could not 
 comprehend those horrors of the mind and soul 
 that Shakespeare and Webster knew and Mar- 
 ston glimpsed. But Tourneur was in sight of the 
 end of greatness ; the period of horrors was com- 
 ing to a close. 
 
 For Beaumont and Fletcher were beginning 
 their fatal reign. At first cleanness and great- 
 ness were still there; and while Beaumont lived 
 the degradation could not go far, for he had a 
 sense of humour and satire. His sentimentality 
 had strength beneath it. He could handle metre 
 like an Elizabethan. None of these things could 
 be said of Fletcher. He had only a kind of wit, 
 a kind of prettiness, and an inelastic sub-variety 
 of the blank verse line. But for the first six years 
 or so, from 1608-1614, they, principally Beau- 
 mont, were doing fairly good work. It is good 
 work of a fatally new kind, but the vices of the 
 new have not yet grown to their full. To these 
 years The Faithful Shepherdess, The Knight of 
 the Burning Pestle, Philaster, and The Maid's 
 
80 JOHN WEBSTER 
 
 Tragedy belong; but drama was on a downhill 
 course. 
 
 It has sometimes been said that the most ex- 
 traordinary gap in the history of our literature, 
 or of any other, is the one between the beginning 
 and the end of the seventeenth century. That 
 little break of twenty years in the middle seems 
 at first sight to have made a tremendous differ- 
 ence. Dry den's inability to understand Shake- 
 speare and his fellows is a commonplace; and 
 one can see how inevitable it was from their 
 minds. The cataclysm of the Civil War, social 
 changes, and the sojourn of the generation 
 abroad, are generally held responsible. (Sir 
 George Etherege saw the premieres of Moliere 
 in Paris.) Closer inspection shows the wrong- 
 ness of this view. Anyone familiar with the life, 
 literature, and drama of court circles just before 
 the outbreak of the Civil War, will realise that 
 the extraordinary thing is how like they are to 
 the products of the Restoration period. There 
 was no gap. Sir John Denham's The Sophy 
 (1641) is almost indistinguishable from a Res- 
 toration play. The true gap is far more remark- 
 able and far earlier. It is hidden by over-lap- 
 pings, but its presence is obvious about the year 
 1611. Five years before that, England was 
 thunderous with the most glorious tragedy and 
 
THE ELIZABETHAN DRAMA 81 
 
 the strangest passion. Five years after that, 
 Fletcher and the silly sweetness of tragi-comedy 
 were all-powerful. The path, unmistakeably the 
 same path, led on and down, through Massinger 
 and Shirley. Five years before that, the intel- 
 lect and the imagination had been dizzily and 
 joyfully up-borne on that wit Chapman thinks 
 so fine : 
 
 "Your wit is of the true Pierian spring, 
 That can make anything of anything." 
 
 It was exhilarating, if sometimes irritating. 
 The wit that succeeded it was Court humour, 
 born of the fancy, touched with softness, 
 feeble-winged. Heart supplanted brain, and 
 senses sense. 
 
 For all this Fletcher was to blame, or, if the 
 causes were deeper, he stands a figurehead for 
 our abuse. What the causes of such movements 
 are, it is always difficult to say. The gradual 
 change in the personnel of the theatre and its au- 
 diences may have had something to do with it. 
 Puritanism and democracy were becoming grad- 
 ually and deplorably identified. This meant that 
 the theatre was being based on only one class. 
 The audiences were becoming upper-class, or 
 of the upper-class party; it is even more note- 
 worthy that the same thing was happening to the 
 
82 JOHN WEBSTER 
 
 dramatists. Henceforward they were almost en- 
 tirely drawn from court circles and the upper 
 classes. Or the reason for the degeneracy may 
 have lain in some deeper weariness of men's 
 hearts. Anyhow, the degeneracy was there. 
 Splendour became softness and tragedy tragi- 
 comedy. These later dramatists were like 
 Ophelia. 
 
 "Thought and affliction, passion, hell itself. 
 She turns to favour and to prettiness." 
 
 It was in this sinking to prettiness and to ab- 
 sence of seriousness that the "degeneracy" of the 
 later Elizabethan drama lies, not, as some mod- 
 ern critics say, in the selection of such admirable 
 subjects as incest for their dramas. Comj^are a 
 typical Fletcherian tragedy, Bonduca, with one 
 of its predecessors. It is the absence of serious 
 intention, the only desire to please, the lack of 
 artistic morality, that make such plays, with 
 their mild jokes, their co-ordinate double plots, 
 and their unreality, so ultimately dreary and 
 fifth-rate to a sensible reader. But such stuff 
 overwhelmed England. That vulgarest of writ- 
 ers, Middleton, who had been doing admirable, 
 coarse, low-level comedy, rather Jonsonian and 
 quite realistic, turned about 1609 to romantic 
 
THE ELIZABETHAN DRAMA 83 
 
 comedy. And by 1612 even Tourneur had writ- 
 ten a tragi-comedy, Tlie Nobleman. 
 
 But even when the triumph of prettiness was 
 on its way to completion, there was one shghtly 
 old-fashioned figure still faithful to that larger 
 prime. Serious tragedy seems only to have 
 reached Webster, after it had left everybody 
 else. In 1612 and 1613 he wrote two of the most 
 amazing products of that amazing period. His 
 powerful personality coloured what he wrote, 
 and yet these two plays are more representative 
 than any that had led to them, of the period be- 
 hind them. The stream swept straight on from 
 Marston and Tourneur to Webster. With him 
 the sinister waves, if they lost something of their 
 strange iridescence, won greater gloom and pro- 
 fundity. After him they plunged into the depths 
 of earth. He stands in his loneliness, first of 
 that long line of "last Elizabethans." As the 
 edge of a cliff seems higher than the rest for 
 the sheer descent in front of it, Webster, the 
 Webster of these two plays, appears even 
 mistier and grander than he really is, because 
 he is the last of Earth, looking out over a sea 
 of saccharine. 
 
Chapter IV 
 
 JOHN WEBSTER 
 
 John Webster is one of the strangest figures 
 in our literature. He was working for quite 
 twenty years. We have at least four plays in 
 which he collaborated, and three by him alone; 
 but through all the period and in all his work 
 he is quite ordinary and undistinguished, except 
 for two plays which come quite close together 
 in the middle. For two or three years, about 
 1612, he was a great genius; for the rest he was, 
 if not indistinguishable, entirely commonplace. 
 Coleridge does not more extraordinarily prove 
 Apollonian fickleness. Webster makes one be- 
 lieve successful art depends as much on a wild 
 chance, a multiple coincidence, as Browning 
 found love did. If he had not had time in that 
 middle period ; if it had come a little later, under 
 the Fletcherian influence; if he had been born 
 twenty years later; if — . . . He was just in 
 time; the subject just suited him; the traditional 
 atmosphere of the kind of play called out his 
 greatest gifts; the right influence had preceded 
 
 84 
 
JOHN WEBSTER 85 
 
 him; he was somehow not free to write the "true 
 dramatic poem" or "sententious tragedy" he 
 wanted to. And so these two great tragedies 
 happened to exist. That easy and comfortable 
 generalisation of the Philistine "genius will out!" 
 finds signal refutation in Webster. I shall give 
 a short general account of his life and activities, 
 and then examine his work more closely. 
 
 We know a great deal about Webster's life. 
 He was born in the latter half of the sixteenth 
 century, and died some time before the end of 
 the seventeenth. He was an Elizabethan 
 dramatist, a friend of Dekker and Chapman and 
 Heywood. He was an odd genius who created 
 slowly and borrowed a great deal. He was not 
 very independent. . . . 
 
 It is, unimportantly, true that fewer "facts" 
 than truths are known about him. We are luck- 
 ily spared the exact dates of his uninteresting 
 birth and death, and his unmeaning address and 
 family. We have not even enough to serve as 
 a frame-work for the elaborate structure of 
 "doubtless" and "We may picture to ourselves 
 young — " that stands as a biography of Shake- 
 speare and others. It could, of course, be done 
 by throwing our knowledge of Elizabethan con- 
 ditions and our acquaintance with the character 
 of the author of The Duchess of Malfi together. 
 
86 JOHN WEBSTER 
 
 It would not be worth it. We know that Web- 
 ster was a member of the Merchant Tailors' Com- 
 pany, and born free of it. There is a late legend 
 that he was clerk of St. Andrew's, Holborn. At 
 one time it seemed possible to identify him (con- 
 temporary enemies tried to) with an ex-army 
 chaplain who wrote fanatical religious tracts and 
 was a University reformer, in the middle of the 
 seventeenth century. Superb thought! It is 
 hard to degenerate nobly; and his contempo- 
 raries, after reaching their summit, went down- 
 hill (as writers) in various ways. Some became 
 dropsical; others entered the Church; others 
 went on writing; a few drank. But this, this 
 would have been an end worthy of a fantastic 
 poet! Alas! Mr. Dyce investigated too thor- 
 oughly, and pretty certainly disproved the iden- 
 tification. After his last play, Webster slips 
 from us inscrutably round the corner. He may 
 have lived on for years and years. He may have 
 died directly. It does not matter to us. 
 
 For the life of Webster the dramatist, how- 
 ever, as opposed to Webster the private man, 
 we have a few facts. He comes into our notice 
 — fairly young, it is to be presumed — in 1602. 
 He was then very busily one of the less important 
 of a band of hack playwrights employed by 
 Henslowe. He had a hand in several plays that 
 
JOHN WEBSTER 87 
 
 we know of during that year : Ccesar's Fall, Two 
 Shapes,^ Christmas comes hut once a year, and at 
 least one part of Lady Jane, His collaborators 
 were Munday, Drayton, Middleton, Heywood, 
 Chettle, Smith, and Dekker. It was the begin- 
 ning, as far as we know, of a close connection 
 with Dekker and a long one with Heywood. 
 Webster was writing for both Henslowe's com- 
 panies, Ccesar's Fall and Two Shapes for the 
 Admiral's men, Christmas comes but once 
 a year and Lady Jane for Worcester's men. 
 Writing for Henslowe was not the best school 
 for genius. No high artistic standard was ex- 
 acted. It rather implies poverty, and certainly 
 means scrappy and unserious work. It may 
 have given Webster — it would have given some 
 people — a sense of the theatre. But he emerged 
 with so little facility in writing, and so little 
 aptitude for a good plot (in the ordinary sense) , 
 that one must conclude that his genius was not 
 best fitted for theatrical expression, into which 
 it was driven. There are other periods and liter- 
 ary occupations it is harder to imagine him in. 
 But I can figure him as a more or less realistic 
 novelist of the present or the last eighty years, 
 preferably from Russia. His literary skill, his 
 
 * Perhaps the same play. See Appendix B. 
 
88 JOHN WEBSTER 
 
 amazing genius for incorporating fragments of 
 his experience, his "bitter flashes" and slow 
 brooding atmosphere of gloom, would have been 
 more tremendous untrammelled by dramatic 
 needs. His power of imaginative visualisation 
 was often superfluous in a play. Like most of 
 his gifts it is literary. It is just what one keenly 
 misses in most novels. One can see, almost quote 
 from, a rather large grey-brown novel by John 
 Webster, a book full of darkly suffering human 
 beings, slightly less inexplicable than Dostoieff- 
 sky's, but as thrilling, figures glimpsed by sud- 
 den flashes that tore the gloom they were part 
 of; a book such that one would remember the 
 taste of the whole longer than any incident or 
 character. . . . But these imaginations are fool- 
 ish in an Heraclitan world, and the phrase "John 
 Webster in the nineteenth century" has no mean- 
 ing. 
 
 Webster seems to have had the ordinary train- 
 ing, collaborating in classical tragedy, history, 
 and low comedy. None of his collaborators left 
 much mark on his style. He was more sub- 
 servient than impressionable. The only play 
 of this lot that we have is Lady Jane, printed 
 in a cut form as Sir Thomas Wyatt. Webster 
 probably had a good deal to do with two Scenes, 
 
JOHN WEBSTER 89 
 
 2 and 16*/ he may be responsible for more, but, 
 if so, it is indistinguishable. The whole play is 
 a ramshackle, primitive (for 1602), ordinary 
 affair. The parts we think Webster's are rather 
 different from the rest, but no better. Metri- 
 cally they are hopeless, but that may be due to 
 the state of the play. There is a sort of sleepy 
 imagination in — 
 
 "Lo, we ascend into our chairs of state, 
 Like funeral coffins^ in some funeral pomp. 
 Descending to their graves !" 
 
 It gratifies one with a feeling of fitness, that 
 Webster should have been thinking of funerals 
 so early as this. Perhaps one is sentimentally 
 misled, and it is really someone else's work. The 
 whole thing is equally uncertain and unimpor- 
 tant. 
 
 The Induction to The Malcontent (1604), 
 our earliest example of Webster's unaided writ- 
 ing, is a slight piece of work, and valueless. The 
 stiff involved sentences are characteristic. The 
 humour is commonplace. It all shows up dully 
 by the rest of the play, which is restive and in- 
 
 ^Sc. 2 is from p. 186, col. 1, "Enter Guildford," to p. 187, 
 " 'cave.' Exeunt." 
 
 Sc. 16 is from p. 199, end, "Enter Winchester," to p. 201, 
 " 'dumb.' Exeunt," 
 
90 JOHN WEBSTER 
 
 flamed with the vigorous, queer, vital, biting style 
 of Marston. 
 
 Webster seems to have gone on in the profes- 
 sion of a hack author. He must have collab- 
 orated in dozens of plays in these years, perhaps 
 written some of his own. He next comes to 
 light writing two comedies of London life with 
 Dekker, Westward Ho (1604) and Northward 
 Ho (1605) . This time it is good work he is con- 
 cerned with, though out of his true line. They 
 were written for the Children of Paul's. Web- 
 ster seems to have been a free-lance at this period, 
 going from company to company. But he must 
 somehow have got a sort of reputation by this 
 time, to be joined with Dekker in this friendly 
 skirmish against Chapman, Jonson, and Mar- 
 ston {Eastward Ho), who were all eminent. 
 And in 1607 it seems to have been worth a pub- 
 hsher's while to put his and Dekker's names 
 on the title-page of Sir Thomas Wyatt, and 
 leave out Chettle, Smith, and Heywood. In 
 Westward Ho and Northward Ho there are a 
 few scenes I think we can be pretty certain are 
 mainly Webster's; Northward Ho, II. 2 and V. 
 1, very probably Westward Ho, I. 1 and III. 
 3, and quite probably Northward Ho, I. 1 and 
 III. 1. One seems to catch a sight of him else- 
 where in the plays; but it is difficult to be cer- 
 
JOHN WEBSTER 91 
 
 tain. In the scenes we attribute to him the sound 
 ^pf a deeper, graver, and duller voice than Dek- 
 Jicr's seems to be heard. It is not altogether 
 fancy. The lightness goes. The bawdy jokes 
 change their complexion a little ; they come more 
 from the heart and less from the pen. The peo- 
 ple in the play do not live any the more or the 
 less, but they become more like dead men and 
 less like lively dolls. The whole thing grows less 
 dramatic; the characters become self-consciously 
 expository — Webster w^as always old-fashioned 
 in this — instead of talking to each other, half- 
 face to us, they turn towards the audience and 
 stand side by side, addressing it. Justiniano's 
 jealousy grows more serious and real when Web- 
 ster takes charge of him, more unpleasantly r^al 
 to himself, and fantastically expressed. And 
 {Northward Ho, II. 2) Mistress Mayberry's 
 sudden disappearance to cry stirs you with an 
 unexpected little stab of pathetic reality not un- 
 like the emotion the later Webster can 
 arouse when he will. But the whole outlines an 
 atmosphere of the plays, and the characters and 
 incidents are far nearer Dekker than Webster. 
 It is only possible to say either that Webster 
 was merely assisting Dekker in these plays, or 
 that his peculiar individuality was either un- 
 grown or dormant. No doubt his romantic clas- 
 
92 JOHN WEBSTER 
 
 sical ideas made him feel he was writing very 
 far down to the public. But he need not have 
 been ashamed, and it may very well have done 
 him good. Good farce is a worthy training for 
 a tragic writer; and these plays are excellent 
 comic farce. The wit is not subtle, the plots 
 have no psychological interest, and the ragging 
 of Chapman is primitive. But the characters 
 have a wealth of vitality, spirits, and comic value. 
 The jokes are often quite good, especially the 
 bawdy ones, and the sequence of events keeps 
 your mind lively and attentive. The general at- 
 mosphere in these two plays has a tang of de- 
 lightful, coarse gaiety, like a country smell in 
 March. They are really quite good, for the 
 rough knock-about stuff they are; among the 
 best in their kind, and that no bad kind. It 
 would be amusing, if it were not so irritating, 
 that many who are authorities in Elizabethan 
 literature are violently and angrily shocked by 
 these two plays, and condemn them as filth. 
 Dr. Ward throws up hands of outraged refine- 
 ment. Professor Schelling has an incredibly 
 funny passage. "They mark the depth of gross 
 and vicious realism to which the comedy of man- 
 ners descended. . . . Some of the figures we 
 would fain believe, in their pruriency and out- 
 spoken uncleanhness of speech, represent an oc- 
 
JOHN WEBSTER 93 
 
 casional aberration, if not an outrageous exag- 
 geration, of the manners of the time. ... In 
 our admiration of the ideal heights at times at- 
 tained by the hterature of the great age of Ehza- 
 beth we are apt to forget that the very amplitude 
 of its vibrations involves an extraordinary range, 
 and that we must expect depths and morasses 
 as well as wholesome and bracing moral heights. 
 . . ." If literary criticism crosses Lethe, and 
 we could hear the comments of the foul-mouthed 
 ghosts of Shakespeare, Marlowe, and Webster 
 on this too common attitude, their outspoken un- 
 cleanliness would prostrate Professor Schelling 
 and his friends. Anger at this impudent attempt 
 to thrust the filthy and degraded standards of 
 the modern middle-class drawing-room on the 
 clean fineness of the Elizabethans, might be ir- 
 relevant in an Essay of this sort. What is 
 relevant is a protest that such thin-lipped writ- 
 ers are not only ridiculous on this point, but 
 also, for all their learning and patience, with- 
 out sufficient authority in Elizabethan literature. 
 It is impossible to trust them. Even in deciding 
 a date, it may be necessary to have sympathy 
 with the Elizabethans. The Elizabethans liked 
 obscenity; and the primness and the wickedness 
 that do not like it, have no business with them. 
 There is a silence of some six years after 
 
94 JOHN WEBSTER 
 
 Northward Ho, We do not know what Web- 
 ster was doing. Somehow he was gaining posi- 
 tion, and preparing himself. In 1611 or 1612 
 he produced The White Devil, the first of the 
 two plays which definitely and uniquely give 
 the world Webster. Last heard of he was a 
 subordinate collaborator; now he is a great, very 
 individual dramatist. The step was enormous; 
 but he had a long time to make it in. If Fate 
 had spared us some of his interim works, we 
 might not be so surprised. 
 
 The preface to The White Devil is important 
 for the light it throws both on Webster and on 
 the general critical ideas of the period. "Evi- 
 demment," says M. Symmes, "Webster dans 
 ce passage est un des premiers a connaitre Fim- 
 portance, le merite, et I'individualite du theatre 
 anglais romantique, comme genre separe." ^ It 
 is too strong. But he does seem to hover in a 
 queer way, between intense pride in his own 
 work and fine appreciation of the best among his 
 contemporaries, and scorn of all these in com- 
 parison to a "true dramatic poem" in the clas- 
 sical style. He shows himself wholly of the 
 Jonson-Chapman school of classicists, in agree- 
 ment with the more cultivated critics. His gloom 
 
 * Symmes: Les Debuts de la Critique Dramatique en Angle- 
 terre, etc. 
 
JOHN WEBSTER 95 
 
 fires up at the imaginary glories of these Satur- 
 nian plays; he is superb in his scorn of his own 
 audience. "Should a man present to such an 
 auditory the most sententious tragedy that ever 
 was written, observing all the critical laws, as 
 height of style, and gravity of person, enrich it 
 with the sententious Chorus, and, as it were, life 
 in death in the passionate and weighty Nuntius ; 
 ..." His arrogance was partly due, no doubt, 
 to pique at the failure of the play and partly 
 to the literary fashion. But it had something 
 natural to him. Even in these plays he so 
 scornfully wrote for the "uncapable multi- 
 tude" of those times there is a sort of classicism. 
 His temperament was far too romantic for it; 
 he was not apt to it, like Chapman. Yet, espe- 
 cially in The White Devil, the unceasing coup- 
 lets at the end of speeches, both in their number 
 and their nature, have a curious archaic effect. 
 One line is connected with the situation, and ex- 
 presses an aspect of it; the next, with the pat 
 expected rhyme, goes to the general rule, and 
 turns the moral. It belonged to Webster's ideal 
 ntemperament in poetry to turn readily and con- 
 jftinually to the greater generalisations. These 
 last lines or couplets always lead out on to them. 
 They went, the classicists, with a kind of glee; 
 
96 JOHN WEBSTER 
 
 they liked to be in touch with permanent vague- 
 nesses. 
 
 Webster's praise of his contemporaries is, how- 
 ever, very discriminating. The order he gives 
 them is instructive: — Chapman; Jonson; Beau- 
 mont and Fletcher; Shakespeare, Dekker, and 
 Heywood. He tells us in this preface, what we 
 could have guessed, that he wrote very slowly. 
 It was natural, as he compiled, rather than com- 
 posed, his plays; working so laboriously from 
 his note-book. He may be imagined following 
 doggedly behind inspiration, glooming over a 
 situation till he saw the heart of it in a gesture 
 or a phrase. He casts the sigh of the confirmed 
 constipate at Heywood and Dekker and Shake- 
 speare for their "right happy and copious in- 
 dustry." His agonies in composition are amus- 
 ingly described in a passage in Fitzjeffry's Notes 
 irom Blachfriars (1620).' 
 
 The White Devil and The Duchess of Malfi 
 rare often described as "revenge-plays," a re- 
 cently-invented genus. Dr. Stoll deals at great 
 length with them in this light, and Professor 
 Vaughan devotes two or three pages of his short 
 essay to summing up the history of the type. 
 There is something in the idea, but not much; 
 and it has been over-worked. To begin with, 
 
 * Given in Dyce's 1857 edition. Introduction, p. xvi. 
 
JOHN WEBSTER 97 
 
 there are far fewer examples of this type than 
 these critics believe. And it is not quite clear 
 what is the thread of continuity they are thinking 
 of. Is it the fact that revenge is the motive in 
 each play? Or is it a special type of play, the 
 criterion of which is its atmosphere, and which 
 generally includes vengeance as a motive? If 
 the second, they must include other plays in their 
 list; if the first, drop some out. The truth is 
 that there is a certain type of play, the plot of 
 which was based on blood-for-blood vendetta, 
 and the atmosphere of which had a peculiar tinge. 
 Kyd started it; it dropped for a bit, and then 
 Marston revived it, rather differently, with great 
 foresight, at an opportune moment. It had a 
 brief boom with Marston, Shakespeare, and 
 Chettle. The atmosphere became indistinguish- 
 able from that of a good many plays of the pe- 
 riod. Tourneur took the atmosphere, and dis- 
 carded the revenge-plot, in The Atheist's Trag- 
 edy, So did The Second Maiden's Tragedy, 
 Chapman happened to take the revenge-motive, 
 and went back to Seneca on his own account. 
 He gives a characteristic account of the meta- 
 physics of the revenge-motive in the Revenge of 
 Bussy,"- Webster used it a little in one of two 
 plays that in other ways resemble the work of 
 
 » Chapman's Tragedies, ed. Parrott, pp. 131-2. 
 
98 JOHN WEBSTER 
 
 other people who used the revenge-plot. That 
 is all. To call The Duchess of Malfi a revenge- 
 play is simply ridiculous. If it is raked in, you 
 must include Othello and a dozen more as well. 
 The whole category is a false one. It would be 
 much more sensible to invent and trace the 
 "Trial-at-law" type, beginning with the Eumen- 
 ides, going down through The Blerchant of Ven- 
 ice, The White Devil, Volpone, The Spanish 
 Curate, and a score more, till you ended with 
 Justice, 
 
 The White Devil and The Duchess of Malfi 
 are so similar in atmosphere that it is sometimes 
 difficult for the moment to remember in which 
 of them some character or speech occurs. But 
 it is convenient to consider them separately; and 
 to take The White Devil first. 
 
 The story is simple. Brachiano conceives a 
 passion for Vittoria, and wins her. She suggests, 
 and he plans, the death of Camillo and Isabella. 
 Their love is discovered by Vittoria's mother, 
 Cornelia. Isabella's brothers, Francisco and 
 Monticelso, try to put an end to it, by giving it 
 rope to hang itself. Before this plan can take 
 effect the murders are committed. Francisco 
 and Monticelso arraign Vittoria for complicity 
 in the murders and for adultery. She is con- 
 demned to imprisonment ; but Francisco, to bring 
 
JOHN WEBSTER 99 
 
 the two nearer final ruin, plots so that she and 
 Brachiano escape together to Padua and marry. 
 Thither he follows them, with some friends, in 
 disguise; and accomplishes their deaths. 
 
 Webster did not handle this tale very skilfully, 
 from the dramaturgic point of view. The play 
 is almost a dramatised narrative. Occasionally 
 the clumsiness of his hand is uncomfortably mani- 
 fest. Generally it does not matter, for his virtues 
 lie in a different aspect of plays from plot-mak- 
 ing. The motives of the various characters are 
 more obscure than they are wont to be in Eliza- 
 bethan plays. On the whole this is a virtue; or 
 seems to be to the modern mind. Characters in 
 a play gain in realism and a mysterious solem- 
 nity, if they act unexplainedly on instinct, like 
 people in real life, and not on rational and pub- 
 licly-stated grounds, like men in some modern 
 plays. 
 
 The play begins with a bang. From the point 
 of view of the plot it is an unusual and unhelpful 
 beginning. Count Lodovico (who turns out later 
 in the play to be an unsuccessful lover of Isa- 
 bella, and who becomes the chief instrument in 
 the downfall of Brachiano and Vittoria) has just 
 been branded. He enters with a furious shout. 
 "Banished!" In this scene there is an instance 
 of a favourite dramatic trick of Webster's, to 
 
100 JOHN WEBSTER 
 
 add liveliness. When some long speech has to 
 be made, where Chapman would give it to one 
 person, Webster divides it between two, con- 
 tinually alternating with a few lines each. It 
 makes the scene "go" in a most remarkable man- 
 ner. In this case Gasparo and Antonelli do it 
 to Lodovico. In The Duchess of Malfi Ferdi- 
 nand and the Cardinal treat the Duchess in this 
 way. 
 
 The next scene introduces the chief characters 
 and the chief emotion. This fatal love, the cause 
 of the whole tragedy, enters most strikingly. 
 Vittoria leaves the stage, Brachiano turns, with 
 a flaming whisper, to Flamineo. He wastes no 
 words. He does not foolishly tell the audience, 
 "I am in love with that woman who has just gone 
 off." 
 
 Brachiano. "Flamineo '* 
 
 Flamineo. "My lord?" 
 Brachiano. "Quite lost, Flamineo." 
 
 A 
 
 Webster thought dramalicaHyr- 
 
 Flamineo, a typicar"taave of Webster's, fills 
 the next few pages with a chorus of quotations 
 from Montaigne. Dramatic is the juxtaposition 
 of the passionate scene between Brachiano and 
 Vittoria, broken by the prophetic Cornelia, the 
 baiting of Brachiano by the Duke and the Cardi- 
 
JOHN WEBSTER 101 
 
 nal, and the pitiful interview of Brachiano and 
 his deserted wife. In the last Webster shews 
 that he can turn to more untroubled tragedy 
 when he wants to : 
 
 "I pray, sir, burst my heart; and in my death 
 Turn to your ancient pity, though not love." 
 
 Rather swiftly, Vittoria (perhaps) and 
 Brachiano, certainly, accomplish the murders; 
 and Vittoria is arrested and tried. The trial 
 scene is prodigiously spirited. There is no hero 
 to enhst our sympathy; it is merely a contest 
 between various unquenchable wickednesses. 
 The rattle of rapid question and answer, sharp 
 with bitterness, is like musketry. Vittoria is 
 wicked; but her enemies are wicked and mean. 
 So one sides with her, and even admires. Her 
 spirit of ceaseless resistance and fury, like the 
 wriggling of a trapped cat, is astonishing. 
 
 "For your names 
 Of whore and murdress, they proceed from you. 
 As if a man should spit against the wind ; 
 The filth returns in's face." 
 
 Flamineo's subsequent affectation of madness 
 and melancholy is made too much of; for the 
 purpose of amusing, perhaps. At this point in 
 the play, the two "villains" part company. Fran- 
 
102 JOHN WEBSTER 
 
 Cisco pursues his way alone. The scene between 
 Brachiano, in his groundless jealousy, and Vit- 
 toria, is tremendous with every kind of beauty 
 and horror; beginning from the extraordinarily 
 iin- Websterian : 
 
 "How long have I beheld the devil in crystal! 
 Thou hast led me, like an heathen sacrifice. 
 With music and with fatal yokes of flowers. 
 To my eternal ruin. Woman to* maiT^ 
 Is either a god or.a^wolf." ^^^ 
 
 The taming of the wild cat, Vittoria, is shown 
 with wonderfully precise and profound psychol- 
 ogy; and all made horrible by the ceaseless and 
 eager prompting of Flamineo. 
 
 "Fie, fie, my lord! 
 Women are caught as you take tortoises; 
 She must be turned on her back." 
 
 The scene of the election of the Pope is an ir- 
 relevant ornament. It is noteworthy that to 
 some extent Webster improved in dramatic craft 
 with time. The Duchess of Malft has fewer such 
 scenes than The White DeviL 
 
 The last part of the play, after it removes to 
 Padua, is one long study of the horror of death. 
 It takes it from every point of view. There is 
 the pathetic incomprehension of Cornelia over 
 young Marcello. "Alas! he is not dead; he is 
 
JOHN WEBSTER 103 
 
 in a trance. Why, here's nobody shall get any- 
 thing by his death. Let me call him again for 
 God's sake." 
 
 There is the difficulty and struggle of the death 
 of so intensely live a man as Brachiano: 
 
 "Oh, thou strong heart! 
 There's such a covenant 'tween the world and it. 
 They're loath to break." 
 
 There is the grotesque parody of death, in 
 Flamineo's 
 
 "Oh I smell soot, 
 Most stinking soot ! The chimney is afire. . . . 
 There's a plumber laying pipes in my guts, it scalds." 
 
 There is the superbness of Vittoria's courage; 
 
 "Yes I shall welcome death 
 As princes do some great ambassadors ; 
 I'll meet thy weapon half-way-" 
 
 There are the "black storm" and the "mist" 
 which drive around Vittoria and Flamineo in the 
 last moments of all. 
 
 The Duchess of Malfi is on the whole a better 
 play than The White Devil. It does not have 
 more of Webster's supreme dramatic moments, 
 but the language is more rich and variously mov- 
 ing — in a dramatic, not merely a literary way. 
 
l< 
 
 JOHN WEBSTER 
 
 It is, even more than The White Devil, in the 
 first half a mere simple narrative of events, lead- 
 ing up to a long-continued and various hell in 
 j the second part. It is often discussed if the plots 
 of The White Devil and The Duchess of Malfi 
 are weak. Webster's method does not really 
 take cognisance of a plot in the ordinary sense 
 of the word. He is too atmospheric. It is like 
 enquiring if there is bad drawing in a nocturne 
 of Whistler's. 
 
 The Duchess of JVIalfi is a young widow, for- 
 bidden by her brothers, Ferdinand and the Car- 
 dinal, to marry again. They put a creature of 
 theirs, Bosola, into her service as a spy. The 
 Duchess loves and secretly marries her steward, 
 Antonio, and has three children. Bosola ulti- 
 mately discovers and reports this. Antonio and 
 the Duchess have to fly. The Duchess is cap- 
 tured, imprisoned, and mentally tortured and 
 put to death. Ferdinand goes mad. In the last 
 Act he, the Cardinal, Antonio, and Bosola are 
 / all killed with various confusions and in various 
 horror. 
 
 The play begins more slowly than The White 
 Devil, Bosola appears near the beginning, and 
 plays throughout a part like that of Flamineo. 
 The great scene in the first Act is the scene 
 of the Duchess's proposal to Antonio. It is full ^ 
 
JOHN WEBSTER 105 
 
 of that perfect, tender beauty which the stormy 
 Webster could evoke when he liked; from the 
 Duchess's preliminary farewell to her maid, 
 
 "Good dear soul, 
 
 Wish me good speed; 
 For I am going into a wilderness 
 Where I shall find nor path nor friendly clue 
 To be my guide." 
 
 to the maid's concluding comment: 
 
 "Whether the spirit of greatness or of women 
 Reign most in her, I know not; but it shows 
 A fearful madness : I owe her much of pity." .^ 
 -^^yJ ' , / ^ ■ . ' ' • ' -^; -'---^•^- • 
 
 There is rather hideous and very typical trag- 
 edy in the scene of Bosola's device to discover 
 the Duchess's secret. The meeting of Bosola and 
 Antonio, at midnight, after the birth of the child, 
 is full of dramatic power and of breathless sus- 
 pense that worthily recalls Macbeth, 
 
 Ant. "Bosola! . . . 
 
 heard you not 
 A noise even now? 
 Bos. From whence ? 
 Ant. From the Duchess's lodging. 
 Bos. Not I : did you ? 
 Ant. I did, or else I dreamed. 
 Bos. Let's walk towards it. 
 Ant. No: it may be 'twas 
 
 But the rising of the wind. 
 Bos. Very likely. . . ." 
 
106 JOHN WEBSTER 
 
 When the news is brought to the brothers that 
 the Duchess has had a child, their anger is hide- 
 ous and, as with passionate people, too imagina- 
 tive. 
 
 After this, and before the events which lead 
 to the catastrophe, that is, between the second 
 and third Acts, there is a long and somewhat 
 clumsy interval. This was rather in the dra- 
 matic fashion of the time. Ferdinand's discovery 
 of the Duchess's guilt breaks finely across a 
 lovely scene of domestic merriment. The plot 
 unravels swiftly. The final parting of the Duch- 
 ess and Antonio is full of a remarkable quiet 
 beauty of phrase and poetry. It is a mere acci- 
 dent that we have discovered that it is entirely 
 composed of fragments of, and adaptations from, 
 Sidney, Donne, Ben Jonson, and others. The 
 scenes of the various tortures of the Duchess 
 form an immense and not always successful sym- 
 phony of gloom, horror, madness, and death. It 
 is only redeemed by the fact that the Duchess can 
 never be quite broken: 
 
 j/^ "I am Duchess of Malii still." 
 
 Only once, just before death, does she let an 
 hysterical cry escape her: 
 
 "any way, for Heaven's sake. 
 So I were out of your whispering." 
 
JOHN WEBSTER 107 
 
 , The superhuman death of the Duchess is finely 
 anti-climaxed by the too human death of Cariola, 
 
 [who fights, kicks, j)rays, and lies. 
 " After the death of the Duchess, there is a 
 slight lull before the rest of the tragedy rises 
 again to its climax. It contains a queer scene of 
 macabre comedy where Ferdinand beats his fan- 
 tastic doctor, and a curious, rather Gothic, ex- 
 jtraneous scene of quietness, where Antonio talks 
 to the echo. The end is a maze of death and 
 madness. Webster's supreme gift is the blind- 
 ing revelation of some intense state of mind at 
 a crisis, by some God-given phrase. All the last 
 half of The Duchess of Malfi is full of them. 
 The mad Ferdinand, stealing across the stage in 
 the dark, whispering to himself, with the dev- 
 astating impersonality of the madman, "Stran- 
 gling is a very quiet death," is a figure one may 
 not forget. And so in the next scene, the too 
 sane Cardinal: — 
 
 "How tedious is a guilty conscience! 
 When I look into the fish-ponds in my garden 
 '/Methinks I see a thing armed with a rake 
 That seems to strike at me." 
 
 It is one of those pieces of imagination one can- 
 not explain, only admire. 
 
 But it is, of course, in or near the moment 
 
108 JOHN WEBSTER 
 
 of death that Webster is most triumphant. He 
 
 •^ adopts the romantic convention, that men are, 
 
 \ in the second of death, most essentially and sig- 
 
 \ "nificantly themselves. In the earher play the 
 
 whole angry, sickening fear of death that a man 
 
 feels who has feared nothing else, lies in those 
 
 terrific words of Brachiano's when it comes home 
 
 to him that he is fatally poisoned: — 
 
 "On pain of death, let no man name death to me: 
 It is a word infinitely terrible." 
 
 Webster knows all the ways of approaching 
 death. Flamineo, with the strange carelessness 
 of the dying man, grows suddenly noble. "What 
 dost think on?" his murderer asks him. 
 
 Flamineo. "Nothing; of nothing; leave thy idle 
 questions. 
 I am i' the way to study a long silence: 
 To prate were idle. I remember nothing. 
 There's nothing of so infinite vexation 
 As man's own thoughts." 
 
 And Webster, more than any man in the world, 
 has caught the soul just in the second of its 
 decomposition in death, when knowledge seems 
 transcended, and the darkness closes in, and 
 boundaries fall away. 
 
 "My soul," cries Vittoria, "like to a ship in a black storm, 
 Is driven, I know not whither." 
 
JOHN WEBSTER 109 
 
 And Flamineo — 
 
 "While we look up to Heaven we confound 
 Knowledge with knowledge, O, I am in a mist." 
 
 So in this play Ferdinand "seems to come to 
 himself," as Bosola says, "now he's so near the 
 bottom." He is still half-mad; but something 
 of the old overweening claim on the universe 
 fires up in the demented brain : 
 
 "Give me some wet hay : I am broken-minded. 
 I do account this world but a dog-kennel: 
 I will vault credit and affect high pleasures 
 , Beyond death." 
 
 For some six years again, after The Duchess 
 of Malfi, we know nothing of Webster's activi- 
 ties. When he comes once more into sight in 
 The Devil's Law-Case (1620) he has shared the 
 fate of the whole drama. It is an attempt to 
 write in the Massinger-Fletcher genus of tragi- 
 comedy. The plot is of so complicated a nature 
 that it would take almost the space of the whole 
 play to set it out fully. Indeed there is scarcely 
 a plot at all, but a succession of plots, interwoven, 
 and each used, in the debased way of that period, 
 almost only to produce some ingeniously start- 
 ling scene, some theatrical paradox. It was, 
 probably, Fletcher who was responsible for this 
 
no JOHN WEBSTER 
 
 love of a succession of dramatic shocks. It suited 
 a part of Webster's taste only too well. 
 
 The main incident of the play is a malicious 
 suit brought by a mother, Leonora, against her 
 son, Romelio, trying to dispossess him on the 
 (false) ground of bastardy. Tacked on to that 
 are various minor affairs, a duel between friends 
 in which both are supposed to have been killed 
 and both marvellously survive, a virgin pretend- 
 ing to be with child, a sick man miraculously 
 cured by an assassin's unintentionally medicinal 
 knife, and so on. The most central incident may 
 have been suggested to Webster by an old play, 
 Lusfs Dominion; the cure he got from a transla- 
 tion of some French yarns. But the question 
 of his originality is unimportant. All his inci- 
 dents aim at that cheap fantasticality which 
 marked this Jacobean drama. And his topics 
 are its well-rubbed coins, romantic friendship, 
 sudden "passion," virginity, duelling, seduction. 
 A most dully debonair world. However, he 
 could not handle them w4th the same touch. 
 Webster stepped the same measures as his con- 
 temporaries, willingly enough — conceitedly even, 
 as his dedication and preface show; but with 
 earlier legs. His characters alternate between 
 being the sometimes charming lay-figures of the 
 time, and wakening to the boisterous liveliness 
 
JOHN WEBSTER 111 
 
 of fifteen years before. Several scenes are very 
 noticeably Jonsonian interludes of farce, sand- 
 wiched between comedy. The vigorous flow of 
 Act II, Scene 1 (pages 114-116) is wholly remi- 
 niscent of the comedy of humours. This is part- 
 ly due to the purely satiric character of some of 
 the passages. The dramatists of the beginning 
 of the century loved to play Juvenal. They 
 would still be railing. Webster was especially 
 prone to it. Repeatedly, in The DeviVs Law- 
 Case, this habit of abuse, directed against one 
 person or the world, recalls Webster's two great 
 plays. There are a score of passages where you 
 immediately cry "Webster!" the note is so indi- 
 vidual. And they are mostly of this satiric kind. 
 Who else could have written (I. 1) : 
 
 "With what a compell'd face a woman sits 
 While she is drawing ! I have noted divers, 
 Either to feign smiles, or suck in the lips, 
 To have a little mouth; ruffle the cheeks 
 To have the dimple seen; and so disorder 
 The face with affectation, at next sitting 
 It has not been the same : . . ." 
 
 The "I have noted" of the professional satirist 
 is unmistakeable. 
 
 But, indeed, the essence of Webster pervades 
 this "tragi-comedy." And the result is that it 
 is as far diiFerent from other tragi-comedies in 
 
112 JOHN WEBSTER 
 
 its spirit, as Measure for Measure is from the 
 comedies among which it is numbered. His vo- 
 cabulary and peculiar use of words peep out 
 on every page; "passionately," "infinitely," 
 "screech-owl," "a lordship," "caroche," "mathe- 
 matical," "dung-hill," "foul" a hundred times; 
 and all in sentences that have the very run of his 
 accents. There are scores of short passages. 
 Webster's characters have the trick of comment- 
 ing on themselves when they are jesting. "You 
 see, my lord, we are merry," cries Romelio (p. 
 Ill), and so Sanitonella (p. 114), "I am merry." 
 The Duchess inevitably comes to one's mind, in 
 that happy moment before her world crumbled 
 about her, "I prithee, when were we so merry?" 
 It is a trick that makes the transience or the un- 
 reality of their merriment stand out against the 
 normal and real gloom. Continually in this play, 
 as in the others, Webster is referring to women 
 painting their faces. The subject had a queer 
 fascination for him. Those other, more obvious, 
 I thoughts of his reappear, too ; his broodings on 
 vdeath and gi^aves. There is the same savagery 
 in his mirth : 
 
 "But do you not think" 
 
 says Jolenta, suddenly, when she has acceded to 
 Romelio's horrible plannings, 
 
JOHN WEBSTER 113 
 
 "I shall have a horrible strong breath now?" 
 RoMELio. "Why?" 
 
 JoLENTA. "O, with keeping your counsel^ 'tis so terrible 
 foul." 
 
 "Bitter flashes" Romelio rightly calls such out- 
 bursts. But he himself achieves wit most suc- 
 cessfully in the same mood and manner. When 
 the Capuchin worries him, before his duel, about 
 religion, he, "very melancholy," retorts with a 
 question about swords — 
 
 "These things, you know," the Capuchin re- 
 plies, "are out of my practice." 
 
 "But these are things, you know, 
 I must practise with to-morrow." 
 
 Romelio sardonically returns. It is very clear 
 throughout that the bitterer Webster's flashes 
 are, the brighter. And in a similar way he livens 
 up when he approaches any emotion such as 
 Jolenta describes, in herself, as "fantastical sor- 
 row." It is the fantastical in emotion or char- 
 acter that inspires him, while the fantastical in 
 situation leaves him comparatively cold. He es- 
 says the latter, dutifully — the usual intellectual 
 paradoxes and morbid conventions of impossible 
 psychology which this kind of drama demanded. 
 In that typically-set Websterian scene (Act III. 
 Scene 3 — A table set forth with two tapers, a 
 
114 JOHN WEBSTER 
 
 death's-head, a book.) between Romelio and Jo- 
 lenta, love, hate, passion, anger, and grief play 
 General Post with all the unnatural speed the 
 Jacobeans loved. He has even invested the starts 
 and turns of the trial-scene with a good deal of 
 interest and much dramatic power. But the an- 
 guish that apes mirth and the mirth that toys 
 with pain wake his genius. He even laughs at 
 himself. You feel an almost personal resentment 
 at being sold, towards the end of the play. Ro- 
 melio's sullen but impressive stoicism is broken 
 by Leonora's entrance with coffins and winding- 
 sheets and that incomparable dirge. 
 
 **. . . Courts adieu, and all delights. 
 All bewitching appetites ! 
 Sweetest breath and clearest eye. 
 Like perfumes, go out and die; 
 And consequently this is done 
 As shadows wait upon the sun. 
 Vain the ambition of kings, 
 Who seek by trophies and dead things 
 To leave a living name behind. 
 And weave but nets to catch the wind/* 
 
 Romelio, like any reader, is caught by the ut- 
 ter beauty of this. He melts in repentance, per- 
 suades his mother, and then the priest, to enter 
 the closet, and then — locks them in with entire 
 callousness and a dirty jest, and goes off to his 
 duel. It is, literally, shocking. But Romelio is 
 
JOHN WEBSTER 115 
 
 one of the two or three characters into whom 
 Webster has breathed a spasmodic life and force. 
 The ordinary dolls of the drama, like Contarino 
 and Ercole, remain dolls in his hands. But the 
 lust and grief of Leonora have some semblance 
 of motion, the suffering of Jolenta has an hys- 
 terical truth, and the figure of Romelio lives 
 sometimes with the vitality of an intruder from 
 another world. He comes out of the earlier 
 drama. He is largely the sort of monster Ben 
 Jonson or Marlowe, or Kyd or Tourneur, or tl^ , 
 earlier Webster likes to picture, malign, immoral, 
 grotesque, and hideously alive. Winifred also 
 is older than 1620. She has an unpleasant vi- 
 vacity, a rank itch of vulgarity, as well as the 
 office of commentator, which reminds one of 
 characters in Webster's two great plays. She 
 is aJBosola in skirts. A sure sign, she grows 
 more excited when love-making is to hand. It _ 
 is typical of Webster that he should smirch with ^ 
 his especial rankness, not only the baser char- 
 acters of this play, but the love-making between 
 his hero and heroine, as he does through Wini- / 
 fred's mouth in the second scene of the play. ^ J 
 Like any Flamineo, she interprets between us 
 and the puppets' dallying, a little disgustingly: 
 
 ''O sweet-breath'd monkeys^ how they grow together !" . . . 
 
116 JOHN WEBSTER 
 
 A few incidents stand out, marked by the 
 darker range of colours of the earlier drama. 
 Contarino's groan that announces that he is not 
 dead (III. 2): 
 
 Con. "Or 
 
 First Surgeon. "Did he not groan?" 
 
 Second Surgeon. "Is the wind in that door still?'* 
 
 has something of the terror and abrupt ghostli- 
 ness of the midnight scene in The Duchess of 
 Malfi (II. 3), or Macbeth, or Jonson's additions 
 to The Spanish Tragedy. And Leonora's mad 
 flinging herself on the ground in III. 3, and ly- 
 ing there, is an old trick that the early Eliza- 
 bethan audiences almost demanded as an essen- 
 tial of Tragedy. It goes back through Ferdi- 
 nand, Bussy, and Marston's heroes, to old Hier- 
 onimo herself. 
 
 Webster's note-book is perhaps a little less 
 apparent in this play than in the two previous. 
 But there are a good many passages we can iden- 
 tify, and a lot more we can suspect. He had 
 fewer "meditations" of the old railing order to 
 compile from his pages of aphorisms and modern 
 instances. But we find repetitions from A Mon- 
 umental Column, The White Devil, and espe- 
 cially The Duchess of Malfi; and Ben Jonson 
 and Sidney have found their way through the 
 
JOHN WEBSTER 117 
 
 note-book into these pages. He still employs 
 soliloquy and the concluding couplet to an extent 
 and in a way that seem queer in a play of this 
 period. But he seems to have become a little 
 more sensible to violent incongruity. He never 
 offends so harshly as he had used. Occasionally 
 still, the stage-machinery creaks loudly enough 
 to disturb the theatrical illusion rather unpleas- 
 antly. Sanitonella is a little abrupt and blunt 
 in exacting information from Crispiano for our 
 benefit: — "But, pray, sir, resolve me, what should 
 be the reason that you . . ."etc. (II. 1). And 
 Romelio's asides are occasionally rather too obvi- 
 ous. In III. 3, when his various proposals to 
 Jolenta have been ineffectual, he is non-plussed ; 
 but only for a second: 
 
 RoMELio (aside) *'This will not do. 
 
 The devil has on the sudden furnished me 
 With a rare charm^ yet a most unnatural 
 Falsehood: no matter, so 'twill take. — " 
 
 But at the end, when everybody reveals who he 
 is, and begins explaining everything that has hap- 
 pened, the tedium of these disentanglings is cut, 
 and the apparently inevitable boredom dodged, 
 by a device that is so audacious in its simplicity 
 as to demand admiration. Leonora, who has ap- 
 parently made good use of her imprisonment 
 
118 JOHN WEBSTER 
 
 in the closet to jot down a jorecis of all the plots 
 in the play, interrupts the growing flood of ex- 
 planations with 
 
 "Cease here all further scrutiny. This paper 
 Shall give unto the court each circumstance 
 Of all these passages !" 
 
 One is too relieved to object. 
 
 Metrically this play is very similar to its two 
 forerunners; though here, as in the handling, 
 Webster seems a little quieter. He is unaffected 
 by the Fletcher influence in metre. The run of 
 his lines is still elusive and without any marked 
 melody, except in one or two passages. The be- 
 ginning lines with w w _ the continual shift- 
 ing and sliding of accent, and the jerky effect 
 of conversation, continue. It was always a blank 
 verse for talking rather than reading. One trick 
 Webster seems to have developed further, the fill- 
 ing out of feet with almost inadequate syllables. 
 Twice in the first five pages "marriage" is a 
 trisyllable. ^'Emotion" fills two feet; and so on. 
 This habit, common between 1580 and 1595, was 
 revived by some writers j^fter 1615. It fits in 
 very queerly with that opposite tendency to the 
 use of trisyllabic feet that Webster greatly in- 
 dulged in. Sometimes the combination is rather 
 piquant. But "marriage" is, perhaps, a symp- 
 
JOHN WEBSTER 119 
 
 torn of an increased steadiness and mastery of 
 rhythm. There are two or three passages where 
 his blank verse is abler and better, in considerable 
 periods, not in short fragments and exclamations, 
 than it had been before. And this is accompa- 
 nied by a greater evenness. Leonora's great 
 speech (III. 3) begins with something of the 
 old ripple: but it dies away: 
 
 ". . . Is he gone then? 
 There is no plague i' the world can be compared 
 To impossible desire; for they are plagu'd 
 In the desire itself. . . . 
 
 O, I shall run mad! 
 For as we love our youngest children best. 
 So the last fruit of our affection, 
 Where-ever we bestow it, is most strong. 
 Most violent, most unresistable. 
 Since 'tis indeed our latest harvest-home. 
 Last merriment 'fore winter. . . ." 
 
 The beauty and pathos of these hues, the com- 
 plete and masterful welding of music and mean- 
 ing, show what fineness is in The DeviVs Law- 
 Case, One could quote many other things as 
 noble, or as admirable, from Romelio's glorious 
 
 **I cannot set myself so many fathom 
 Beneath the height of my true heart, as fear," 
 
 or the sagacious and horrid rightness of his 
 
120 JOHN WEBSTER 
 
 "doves never couple without 
 A kind of murmur/' 
 
 to Jolenta's cry, 
 
 "O, if there be another world i' the moon 
 As some fantastics dream. . . ." 
 
 Yet the play is not a good play. These good 
 bits illuminate, for the most part, nothing but 
 themselves, and have only a literary value. A 
 good play must leave an increasing impression 
 of beauty or terror or mirth upon the mind, 
 heaping its effect continually with a thousand 
 trifles. This does not so. It is a play without 
 wholeness. Its merits are occasional and acci- 
 dental. If you read closely, there is the ex- 
 traordinary personality of Webster plain enough 
 over and in it all. But he was working in an 
 uncongenial medium. It is a supreme instance 
 of the importance of the right form to the artist. 
 The Fletcher-Massinger "tragi-comedy" was 
 the product of an age and temper as unsuitable 
 to Webster as the tragedy of blood and dirt had 
 been suitable. The Devil's Law-Case is not even 
 a fine failure, as, for instance, Timon of Athens 
 is. In the first place a tragi-comedy is not a 
 thing to make a fine failure of. And in the sec- 
 ond place Webster's nature and methods de- 
 manded success in a right form, or nothing. He 
 
JOHN WEBSTER 121 
 
 had to suffuse the play with himself. He was 
 not great enough and romantic enough to con- 
 fer immortality upon fragments. His bitter 
 flashes required the background of thunderous 
 darkness to show them up ; against this grey day- 
 light they are ineffectual. 
 
 Beyond the uninteresting and unimportant A 
 Monumental Column (1613), which only shows 
 how naturally Webster turned to the imitation 
 of Donne when he turned to poetry, the uncer- 
 tain and featureless Monuments of Honour, and 
 a few rather perfunctory verses of commenda- 
 tion, we have nothing more of Webster's except 
 A Cure for a Cuckold, This must have been 
 written shortly after The Devil's Law-Case, It 
 is almost entirely unimportant for throwing light 
 on the real Webster. All we know is that he 
 had something to do with the play; how much 
 or little it is impossible to tell from reading it. 
 He may be responsible for the whole of the main 
 plot. That it is not so obscure and unmotivated 
 as has sometimes been supposed, I have shown 
 in an Appendix ; but it is not good. Parts have a 
 slight, unreal, charm for those who are interested 
 in antiquities. The way in which in IV. 3 (p. 
 310) Lessingham suddenly sulks, and goes off 
 
122 JOHN WEBSTER 
 
 to make mischief, in order to spin the play out 
 for another act and a bit, is childish. 
 
 It is a pity we cannot barter with oblivion and 
 give A Cure for a Cuckold for Ford and Web- 
 ster's lost murder play. This was one of the 
 last, and it must have been one of the best, of 
 the Elizabethan domestic tragedies. What a 
 superb combination. Ford and Webster! And 
 on such a subject! It may have been again, after 
 all those years, the last cry of the true voice of 
 Elizabethan drama. Once, in 1624, there was, 
 perhaps, a tragedy of blood, not of sawdust. 
 It is beyond our reach. 
 
Chapter V 
 
 SOME CHARACTERISTICS OF 
 WEBSTER 
 
 It happens, with some writers, that when you 
 come to examine their less-known works, your 
 idea of them suffers considerable change, and 
 you realise that the common conception of them 
 is incomplete, distorted, or even entirely wrong. 
 This is not the case with Webster. He is known 
 to everyone by two plays — The Duchess of Malfi 
 and The White Devil The most diligent study 
 of the rest of his authentic works will scarcely 
 add anything of value to that knowledge of him. 
 jHe is a remarkable dramatist, with an unusually 
 [individual style and emotional view of the world. 
 What "Webster," the literary personality, means 
 to us, its precise character, and its importance, 
 can be discovered and explained from these two 
 plays. So I shall chiefly consider and quote them, 
 with an occasional sidelight from The Devil's 
 Law-Case. 
 
 It is one task of a critic, no doubt, to communi- 
 cate exactly his emotions at what he is criticising, 
 
 123 
 
124 JOHN WEBSTER 
 
 to express and define the precise savour. But it 
 is not a thing one can go on at for long. Hav- 
 ing tried to hint once or twice what "Webster" 
 precisely is, I had better analyse various aspects 
 of him, and not tiresomely, like some political 
 speaker, seek about for a great many ways of 
 saying the same thing. And after all, Webster 
 carries his own sense and savour. A showman, 
 "motley on back and pointing-pole in hand," can 
 but draw attention, and deliver a prologue. If 
 I can explain briefly to anyone the sort of plays 
 Webster was writing, the sort of characters that 
 he took delight in, the kind of verse he used, the 
 kind of literary effect he probably aimed at — 
 as I see all these things — I can then only take 
 him up to a speech of the Duchess and leave him 
 there. One cannot explain 
 
 "What would it pleasure me to have my throat cut 
 With diamonds? or to be smothered 
 With cassia ? or to be shot to death with pearls ? 
 I know death hath ten thousand several doors 
 For men to take their exits ; and 'tis found 
 They go on such strange geometrical hinges 
 You may open them both ways : . . ." 
 
 To paraphrase it, or to hang it with epithets, 
 would be silly, almost indecent. One can only 
 quote. And though quotation is pleasant, it is 
 a cheap way of filling space; and I have written 
 
CHARACTERISTICS 125 
 
 this essay on the assumption that its readers 
 will be able to have at least TJie Duchess of Malfi 
 and The White Devil before them. 
 
 So I shall only attempt, in this chapter, to men- 
 tion some of Webster's most interesting charac- 
 teristics, and to analyse one or two of them. 
 
 His general position, as the rearguard of the 
 great period in Elizabethan drama and Htera- 
 ture, I have already outlined. He took a certain 
 kind of play, a play with a certain atmosphere, 
 which appealed to him, and made two works of 
 individual genius. Beyond this type of play 
 and the tradition of it, there are no very im- 
 portant "influences" on him. Shakespeare's 
 studies of madness may have affected him. The 
 Duchess, 
 
 "I'll tell thee a miracle; 
 I am not mad yet, to my cause of sorrow; 
 The heaven o'er my head seems made of molten brass, 
 The earth of flaming sulphur, yet I am not mad," 
 
 has a note of Lear in it, but also, and perhaps 
 more definitely, of Antonio and Mellida. From 
 Ben Jonson and Chapman he borrowed. And 
 something of their attitude to drama became his. 
 But he does not imitate them in any important 
 individual quality. He pillaged Donne, too, as 
 much of him as was accessible to a middle-class 
 dramatist, and occasionally seems to emulate the 
 
126 JOHN WEBSTER 
 
 extraordinary processes of that mind. The char- 
 acters in Webster's plays, hke the treatment of 
 the story, in as far as they are not his own, are 
 the usual characters of the drama of eight years 
 before. Once only does he noticeably seem to 
 take a figure from the popular gallery of the 
 years in which he was writing. The little prince 
 Giovanni, like Shakespeare's Mamillius, is 
 adopted from the Beaumont and Fletcher chil- 
 dren. He has the same precocity in wit (it seems 
 a httle distressing to modern taste), and more 
 of their sentimentality than Hermione's son. 
 But, against that background, he is, on the whole, 
 a touching and lovely figure. 
 
 The one influence upon Webster that is al- 
 ways noticeable is that of satire. His nature 
 tended to the outlook of satire; and his plays 
 give evidence that he read Elizabethan, and in 
 some form Latin satire with avidity. Hamlet, 
 the Malcontent, and all the heroes of that type 
 of play, "railed" continually. But with Webster 
 every character and nearly every speech has 
 something of the satirical outlook. They de- 
 scribe each other satirically. They are for ever 
 girding at the conventional objects of satire, cer- 
 tain social follies and crimes. There are several 
 little irrelevant scenes of satire, like the malevo- 
 lent discussion of Count Malatesti {D.M,, III. 
 
CHARACTERISTICS 127 
 
 3). It is incessant. The topics are the ordinary 
 ones, the painting of women, the ingratitude of 
 princes, the swaggering of blusterers, the cow- 
 ardice of pseudo-soldiers. It gives part of the 
 peculiar atmosphere of these plays. 
 
 This rests on a side of Webster's nature, which, 
 in combination with his extraordinary literary 
 gifts, produces another queer characteristic of his , 
 — his fondness for, and skill in comment. HeJ 
 is rather more like a hterary man trying to write' 
 for the theatre than any of his contemporaries. 
 Theatrically, though he is competent and some- 
 times powerful, he exhibits no vastly unusual 
 ability. It is his comments that bite deep. Such 
 gems as Flamineo's description of Camillo: 
 
 "When he wears white satin one would take him 
 by his black muzzle to be no other creature than a 
 maggot;" 
 
 or of the Spanish ambassador : 
 
 "He carries his face in's ruff, as I have seen a 
 serving man carry glasses in a cipress hat-band, 
 monstrous steady, for fear of breaking: he looks 
 like the claw of a black-bird, first salted, and then 
 broiled in a candle;" 
 
 or Lodovico's of the black woman Zanche in love : 
 
 "Mark her, I prithee ; she simpers like the suds 
 A collier hath been washed in;" 
 
128 JOHN WEBSTER 
 
 have frequently been quoted. They have a 
 purely literary merit. In other places he 
 achieves a dramatic effect, which would be a lit- 
 tle less in a theatre than in the book, by com- 
 ment. When Bosola brings the terrible discov- 
 ery of the secret to Ferdinand and the Cardinal, 
 he communicates it to them, unheard by us, up- 
 stage. We only know, in reading, how they take 
 it, by the comments of Pescara, Silvio, and Delio, 
 who are watching, down-stage — 
 
 Pesc. "Mark Prince Ferdinand: 
 
 A very salamander lives in's eye, 
 
 To mock the eager violence of fire." 
 SiL. "That cardinal hath made more bad faces with his 
 
 oppression than ever Michael Angelo made good 
 
 ones: he lifts up's nose like a foul porpoise before 
 
 a storm." 
 Pes. "The Lord Ferdinand laughs." 
 Del. "Like a deadly cannon 
 
 That lightens ere it smokes . . ." 
 
 it goes straight to the nerves. "The Lord Ferdi- 
 nand laughs." It is unforgettable. 
 
 Webster had always, in his supreme moments, 
 that trick of playing directly on the nerves. It 
 is the secret of Bosola's tortures of the Duchess, 
 and of much of Flamineo. Though the popular 
 conception of him is rather one of immense gloom 
 and perpetual preoccupation with death, his 
 power lies almost more in the intense, sometimes 
 
CHARACTERISTICS 129 
 
 horrible, vigour of some of his scenes, and his 
 uncanny probing to the depths of the heart. In 
 his characters you see the instincts at work jerk- 
 ing and actuating them, and emotions pouring 
 out irregularly, unconsciously, in floods or spurts 
 and jets, driven outward from within, as you 
 sometimes do in real people. 
 
 The method of progression which Webster 
 used in his writing, from speech to speech or 
 idea to idea, is curiously individual. The ideas 
 do not develop into each other as in Shakespeare, 
 nor are they tied together in neatly planned 
 curves as in Beaumont and Fletcher. He seems 
 to have, and we know he did, put them into the 
 stream of thought from outside ; plumping them 
 down side by side. Yet the very cumbrousness 
 of this adds, in a way, to the passion and force 
 of his scenes, as a swift stream seems swifter and 
 wilder when its course is broken by rocks and 
 boulders. The craft of Shakespeare's genius 
 moves with a speedy beauty like a yacht running 
 close into the wind ; Websters is a barge quanted' 
 slowly but incessantly along some canal, cum- 
 brous but rather impressive. 
 
 This quality of the progression of Webster's 
 thought, and, in part, of his language, contrasts 
 curiously with his metr6. The Elizabethan use of 
 blank verse was always liable to be rather fine; 
 
130 JOHN WEBSTER 
 
 but there was only a short period, and it was 
 only in a few writers, that it got really free — 
 until its final dissolution in the thirties. Web- 
 ster was one of these writers, probably the freest. 
 Only Shakespeare can approach him in the lib- 
 erties he took with blank verse ; but Shakespeare's 
 liberties conformed to higher laws. Webster 
 
 ^probably had a worse ear for metre, at least in 
 blank verse, than any of his contemporaries. His 
 
 I verse is perpetually of a vague, troubled kind. 
 Each line tends to have about ten syllables and 
 about five feet. It looks in the distance like a 
 blank verse line. Sometimes this line is extraor- 
 dinarily successful ; though it is never quite scan- 
 nable. Brachiano's 
 
 "It is a word infinitely terrible,** 
 
 is tremendously moving. But sometimes Web- 
 ster's metrical extravagance does not justify it- 
 self, and rather harasses. The trick of beginning 
 a line with two unaccented syllables, if repeated 
 too often in the same passage, does more to break 
 the back of the metre than almost any other pos- 
 sible peculiarity. 
 
 On the whole it is probable that Webster did 
 all this on purpose, seeing that a larger licence 
 of metre suits blank verse in drama than is per- 
 missible in literature. When he turned poet, 
 
CHARACTERISTICS 131 
 
 in A Monumental Column, he is equally unmet- 
 rical; but that, can probably be attributed to 
 the very strong influence of Donne. Certainly 
 the lyrics in his plays would seem to show that 
 as a lyric poet he could have been among the 
 greatest, a master of every subtlety, at least of 
 that lyric metre which he did use. It is the one 
 which the Elizabethans, almost, invented, and 
 upon which they performed an inconceivable va- 
 riety of music. Milton, who learnt so much from 
 them in this respect, made this metre the chief 
 part of his heritage. But even he could not in- 
 clude all that various music. It is the metre of 
 L' Allegro, II Penseroso, and the end of Comus, 
 No man ever got a stranger and more perfect 
 melody from it than Webster in his dirges. 
 
 Webster's handling of a play, and his style of 
 writing, have something rather slow and old- 
 fashioned about them. He was not like Shake- 
 speare or Beaumont and Fletcher, up-to-date and 
 "slick." He worried his plays out with a grunt- 
 ing pertinacity. There are several uncouth char- 
 acteristics of his that have an effect which halts 
 between archaism and a kind of childish awk- 
 wardness, like "primitive" art of various nations 
 and periods. Sometimes he achieves the same 
 result it can have, of a simplicity and directness 
 refreshingly different from later artifice and ac- 
 
132 JOHN WEBSTER 
 
 complishment. Sometimes he only seems, to the 
 most kindly critic, to fail hopelessly for lack of 
 skill. One of these characteristics is the use of 
 couplets, usually to end the scene, and commonly 
 of a generalising nature. This is, of course, old- 
 fashioned. The frequency of such couplets is 
 an often-noticed feature of the early Elizabethan 
 drama: and the plays of such a writer as Shake- 
 speare are dated by the help of the percentage 
 of rhyming to unrhyming lines. Even as late 
 as Webster, other authors sometimes ended the 
 play, or a scene, with a couplet. But they did 
 it with grace; using it almost as a musical de- 
 vice, to bring the continued melody of their verse 
 to a close. And in the earlier plays, where one 
 or more rhyming couplets end most scenes and 
 many speeches, and even, especially in the more 
 lyrical parts, come into the middle of passages, 
 the rest of the versification is of a simple, rhyth- 
 mical end-stopped kind; and so the couplets 
 seem scarcely different from the rest, a deeper 
 shade of the same colour. Webster's couplets 
 are electric green or crimson, a violent contrast 
 with the rough, jerky, sketchy blank verse he 
 generally uses. Some of them are so incongru- 
 ous as to be ridiculous. At the end of a stormy 
 passage with the Cardinal, Ferdinand says : 
 
CHARACTERISTICS 133 
 
 "In, in; I'll go sleep. 
 Till I know who leaps my sister, I'll not stir; 
 That known, I'll find scorpions to sting my whips. 
 And fix her in a general eclipse." [Exeunt. 
 
 If you consider the general level of Webster's 
 writing, this rings almost childish. In llie 
 White Devil there are two instances of rhyming 
 couplets close to each other, one superbly suc- 
 cessful, the other a failure. The rather hideous 
 and queerly vital wooing-scene between Brachi- 
 ano and Vittoria leads up to a speech of the 
 former's that ends: 
 
 "You shall to me at once, 
 Be dukedom, health, wife, children, friends, and all." 
 
 Cornelia, Vittoria's mother, who has been listen- 
 ing behind, unseen, breaks the tension with a 
 rush forward and the cry: 
 
 "Woe to light hearts, they still forerun our fall!" 
 
 It has a Greek ring about it. It brings the fresh 
 and terrible air of a larger moral world into the 
 tiny passionate heat of that interview. And 
 withal there is a run of fine music in the line. 
 The rhyme helps all this materially. It enhances 
 and marks the moment, and assists the play. But 
 a dozen lines later, after some burning speeches 
 of reproach in ordinary blank verse, Corneha 
 
 1 
 
134 JOHN WEBSTER 
 
 drops into rhyme again to show the moral of 
 it all: 
 
 "See, the curse of children ! 
 In life they keep us frequently in tears; 
 And in the cold grave leave us in pale fears." ^ 
 
 The end of the play affords even more ex- 
 traordinary examples of these couplets. Sand- 
 wiched in between the dying Vittoria's tremen- 
 dous 
 
 "My soul, like a ship in a black storm. 
 Is driven, I know not whither," 
 
 and Flamineo's equally fine sentence — an exam- 
 ple of generalisation rightly and nobly used — 
 
 "We cease to grieve, cease to be fortune's slaves. 
 Nay, cease to die, by dying," 
 
 comes the smug and dapper irrelevancy of 
 
 "Prosperity doth bewitch men, seeming clear; 
 But seas do laugh, show white, when rocks are near." 
 
 It is beyond expression, the feeling of being let 
 down, such couplets give one. 
 
 In three places a diif erent and very queer side 
 of Webster's old-fashionedness or of his occa- 
 sional dramatic insensibility, is unpleasantly 
 
 *This couplet seems even absurdcr to us than it should, because 
 the word "frequently" has since Webster got a rapid colloquial 
 sense of "quite often." 
 
CHARACTERISTICS 135 
 
 manifest. Here it becomes plainer, perhaps, that it 
 is rather a childish than an old-fashioned tendency 
 which betrays him to these faults. Three times, 
 once in The White Devil, and twice in The Duch- 
 ess of Malfi, the current of quick, living, realistic 
 speeches — each character jerking out a hard, bit- 
 ing, dramatic sentence or two — is broken by long- 
 winded, irrelevant, and fantastically unrealistic 
 tales. They are of a sententious, simple kind, 
 such as might appear in ^Esop. Generally they 
 seem to be lugged in by their ears into the play. 
 They are introduced with the same bland, start- 
 ling inconsequence with which some favourite 
 song is brought into a musical comedy, but with i 
 immeasurably less justification. The instance in / 
 The White Devil is less bad than the others." 
 Francisco is trying to stir Camillo against the 
 indignity of horns. He suddenly tells him a long 
 tale how Phoebus was going to be married, and 
 the trades that don't like excessive heat made 
 a deputation to Jupiter against the marriage, 
 saying one sun was bad enough, they didn't want 
 a lot of little ones. So, one Vittoria is bad 
 enough ; it is a good thing there are no children. 
 It is pointless and foolish enough, in such a play. 
 But the instances in The Duchess of Malfi sur- 
 pass it. In the tremendous scene in the bed- 
 chamber when Ferdinand accuses the Duchess 
 
136 JOHN WEBSTER 
 
 of her marriage, the mad frenzy of his reproaches 
 is excellently rendered. She replies with short 
 sentences, bursting from her heart. Each of his 
 taunts carries flame. The whole is hving, terse, 
 and affecting. In the middle of this Ferdinand 
 breaks into a long old-fashioned allegory about 
 Love, Reputation, and Death, a tale that (but 
 for a fine line or two) might have appeared in 
 any Elizabethan collection of rhymed parables. 
 The point of it is that Reputation is very easy 
 to lose, and the Duchess has lost hers. It is as 
 irrelevant and not so amusing as it would be if 
 Michael Angelo had written a Christmas cracker 
 posy on the scroll the Cumaean Sibyl holds. In 
 the third instance the Duchess mars the end of 
 a lovely and terrible scene (III. 5) by a would- 
 be funny moral tale about a dogfish and a sal- 
 mon. Here there is a sort of pathetic suitability 
 in the Duchess, half broken with sorrow^ almost 
 unconsciously babbling childish tales to her ene- 
 mies. But, with the other tales in mind, one 
 finds it hard to believe Webster meant this. If 
 he did, he did not bring his effect off. The tale 
 is too incongruous with the rest of the scene. 
 
 There are still further instances of Webster's 
 occasional extraordinary childishness in drama, 
 namely his shameless use of asides, soliloquies, 
 and other devices for telling his audience the 
 
CHARACTERISTICS 137 
 
 motives of the actors or the state of the plot. 
 The Elizabethans were always rather careless. 
 The indiscriminate soliloquy or aside were part 
 of their inheritance, which they but gradually got 
 rid of. If soliloquies, and even asides, are 
 handled rightly, in a kind of drama like the 
 Elizabethan, they need not be blemishes. They 
 can add greatly to the play. Hamlet's solilo- 
 quies do. The trend of recent dramatic art has 
 been' unwise in totally condemning this stage de- 
 vice. There are two quite distinct effects of 
 soliloquy in a play. One is to tell the audience 
 the plot; the other is to let them see character or 
 feel atmosphere. The first is bad, the second 
 good. It is perfectly easy for an audience to ac- 
 cept the convention of a man uttering his 
 thoughts aloud. It is even based on a real occur- 
 rence. When the man is alone on the stage it 
 is an entirely simple and good convention. Even 
 if there are other characters present, i.e. when the 
 soliloquy approaches the aside, the trick only 
 needs careful artistic handling. But the essen- 
 tial condition is that the audience feels it is over- 
 hearing the speaker, as much, at least, as it over- 
 hears the dialogue of the play. In soliloquies 
 or in dialogues the characters may, to a certain 
 extent, turn outward to the audience, and ad- 
 dress them ; in the same way as they forbear from 
 
138 JOHN WEBSTER 
 
 often turning their backs on them. But solilo- 
 quies must go no further. So far, they are ac- 
 ceptable. If we can accept the extraordinary 
 convention that a man's conversation shall be 
 coherent, and in blank verse to boot, we can 
 easily swallow his thoughts being communicated 
 to us in the same way. It is only when the 
 dramatist misuses this licence, and foists improb- 
 able and unnaturally conscious thoughts on a 
 man, in order to explain his plot, that we feel 
 restive. The fault, of course, lies in the unnat- 
 uralness and the shameless sudden appearance 
 of the dramatist's own person, rather than in 
 the form of a soliloquy. Only, soliloquies are es- 
 pecially liable to this. A legitimate and superb 
 use of soliloquy occurs near the end of The 
 Duchess of Malfi, in a passage from which I 
 have already quoted, where the Cardinal enters, 
 alone, reading a book: 
 
 "I am puzzled in a question about hell: 
 He says^ in hell there's one material fire, 
 And yet it shall not burn all men alike, 
 Lay him by. 
 
 — How tedious is a guilty conscience! 
 When I look into the fish-pond in my garden, 
 Methinks I see a thing arm'd with a rake. 
 That seems to strike at me." 
 [Enter Bosola and Servant hearing Antonio's hodyJ] 
 
CHARACTERISTICS 13& 
 
 This is an entirely permissible and successful 
 use of soliloquy. The words and thought are 
 mysteriously thrilling. They sharpen the agony 
 of the spectator's mind to a tense expectation; 
 which is broken by the contrast of the swift pur- 
 pose of Bosola's entry, with the sei^ant and the 
 body, and the violent progression of events en- 
 suing. The whole is in tone together; and the 
 effect bites deep, the feeling of the beginning of 
 sheeting rain, breaking the gloomy pause before 
 a thunderstorm. But there are cases of Webster 
 using the soliloquy badly. In The White Devil, 
 when the servant has told Francisco that Brachi- 
 ano and Vittoria have fled the city together, he 
 goes out. Francisco is left alone, exclaiming, 
 "Fled! O, damnable!" He immediately alters 
 his key: 
 
 "How fortunate are my wishes ! Why, *twas this 
 I only laboured! I did send the letter 
 To instruct him what to do," etc., etc. 
 
 One finds the dramatist rather too prominently 
 and audibly there. But his presence becomes 
 even more offensive when he is visible behind two 
 characters and their dialogue, as in the instance 
 from The DeviVs Law-Case, II. 1. A worse 
 case of this, both in itself and because it comes in 
 a tragedy, occurs in The White Devil, where 
 
140 JOHN WEBSTER 
 
 Francisco and Monticelso explain their actions 
 to each other, after Camillo, charged with the 
 commission against the pirates, has made his 
 exit. 
 
 Francisco. "So, 'twas well fitted: now shall we discern 
 
 How his wish'd absence will give violent way 
 
 To Duke Brachiano's lust." 
 Monticelso. "Why, that was it; 
 
 To what scorned purpose else should we make choice 
 
 Of him for a sea-captain.''" etc. 
 
 But having informed us of their motives in 
 this, Webster suddenly remembers that we may- 
 say, "But why should they start on such a line 
 of action at all?" So Monticelso, later in the 
 conversation, apropos of nothing in particular^ 
 remarks — 
 
 "It may be objected, I am dishonourable 
 To play thus with my kinsman; but I answer, 
 For my revenge I'd stake a brother's life. 
 That, being wrong'd, durst not avenge himself." 
 
 A very similar instance of a pathetic attempt 
 •to make the audience swallow the plot, by care- 
 jfully explaining the motives, is in the fourth act 
 of The Duchess of Malfi, a play distinctly less 
 disfigured by these childishnesses of Webster's 
 than The White Devil. There Ferdinand, in 
 what purports to be a conversation with Bosola, 
 goes back in his mind and rakes out, all unasked, 
 
CHARACTERISTICS 141 
 
 his two motives for persecuting the Duchess. 
 His behaviour, though badly portrayed, is less 
 unconvincing and improbable than The White 
 Devil instance. But such blunders make even 
 the asides of Flamineo, when he is explaining his 
 antic behaviour to the audience, flagrant as they 
 are, seem mild and legitimate stage-devices. 
 
 A special class of unrealistic asides and con- 
 versations, and one very much affected by the 
 Elizabethans, is the situation when A., B., and 
 C. are on the stage, and B. and C. are carrying 
 bn a conversation, interspersed with asides be- 
 tween A. and B. which C. does not notice. Peo- 
 ple who have experience of the stage know how 
 almost impossible this is to manage with any 
 show of probability. In a comedy or farce the 
 absurdity matters less. But the scene between 
 Lodovico, Francisco, and Zanche, after Brachi- 
 ano's death, though it partakes of farce, makes 
 one uneasy. 
 
 All these childishnesses and blunders in Web- 
 ster's plays, soliloquies, asides, generalisations, 
 couplets, and the rest, are due, no doubt, to care- 
 lessness and technical incapacity. His gifts were 
 of a different kind. But the continual general- 
 isations arise also from a particular bent of his 
 mind, and a special need he felt. It is normal in 
 
142 JOHN WEBSTER 
 
 the human mind, it was unusually strong in the 
 Elizabethans, and it found its summit in Webster 
 of all of that time — the desire to discover the 
 general rule your particular instance illustrates, 
 and the delight of enunciating it. Many people 
 find their only intellectual pleasure in life, in the 
 continued practice of this. But drama seems, 
 or seemed, to demand it with especial hunger; 
 most of all the poetic drama. The Greeks felt 
 this, and in the form of drama they developed 
 this was one of the chief intellectual functions of 
 the chorus. I say "intellectual," meaning that 
 in their music and movement they appealed 
 through other channels to the audience — though 
 here, too, in part, to something the same taste 
 in the audience, that is to say, the desire to feel 
 a little disjunct from the individual case, and 
 to view it against some sort of background. 
 Metre itself has, psychologically, the same effect, 
 a little. But the brain demands to be told ^o m^ 
 
 ffyvpaL vLKo. or fxlfivet. Se jjiifxpovro^ ly xpov^ff Aios iro-deiv rov epiavTa^ 
 
 or any of the other deductions and rules. 
 
 The Greeks, then, received, to their satisfac- 
 tion, the knowledge of other instances or of the 
 general rule or moral, from the chorus. It is 
 interesting to see the various ways of achieving 
 the effects of a chorus that later drama has used. 
 For to some extent the need is always felt, 
 
CHARACTERISTICS 143 
 
 though not violently enough to overcome the 
 dramatic disadvantages of an actual chorus. 
 Sometimes one character in a play is put aside 
 to serve the purpose, like the holy man in Max- 
 im Gorki's The Lower Depths. Or the char- 
 acters sit down and, a little unrealistically, argue 
 out their moral, as in Mr. Shaw's plays. Mr. 
 Shaw and a good many modern German, Eng- 
 lish, and Scandinavian writers, also depend on 
 the spectator having picked up, from prefaces 
 and elsewhere, the general body of the author's 
 views against the background of which any par- 
 ticular play is to be performed. Ibsen had two 
 devices. One was to sum up the matter in some 
 prominent and startling remark near the end, 
 like the famous "People don't do such things!" 
 The other was to have a half-mystical back- 
 ground, continually hinted at; the mountain- 
 mines in John Gabriel Borkman, the heights in 
 When We Dead Awaken, the sea in The Lady 
 from the Sea, the wild duck. In certain catch- 
 words these methods met; "homes for men and 
 women," "ghosts," "you don't mean it!" and the 
 rest. The temptation to point a moral in the 
 last words of a play is almost irresistible; and 
 sometimes justified. A well-known modern play 
 called Waste ends, "the waste! the waste of it 
 all!" The Elizabethans were very fond of doing 
 
144 JOHN WEBSTER 
 
 this. They had the advantage that they could 
 end with a rhymed couplet. But they were liable 
 to do it at the end of any scene or episode. It 
 has been pointed out how much Webster was 
 addicted to this practice. Towards their close 
 his plays became a string of passionate generali- 
 ties. Antonio and Vittoria both die uttering 
 warnings against "the courts of princes." Other 
 characters alternate human cries at their own 
 distress with great generalisations about life and 
 death. These give to the hearts of the spectators 
 such comfort and such an outlet for their con- 
 fused pity and grief as music and a chorus afford 
 in other cases. But Webster also felt the need 
 of such broad moralising in the middle of his 
 tragedies. Sometimes he pours through the 
 mouth of such characters as Bosola and Fla- 
 mineo, generalisation after dull generalisation, 
 without illuminating. Greek choruses have 
 failed in the same way. But when a gnome that 
 is successful comes, it is w^orth the pains. The 
 solidity and immensity of Webster's mind behind 
 /the incidents is revealed. Flamineo fills this part 
 at the death of Brachiano. But often he and 
 Bosola are a different, and very Websterian, cho- 
 rus. Their ceaseless comments of indecency and 
 mockery are used in some scenes to throw up by 
 contrast and enhance by interpretation the pas- 
 
CHARACTERISTICS 145 
 
 sions and sufferings of human beings. They pro- 
 vide a background for Prometheus ; but a back- 
 ground of entrails and vultures, not the cliffs 
 of the Caucasus. The horror of suffering is in- 
 tensified by such means till it is unbearable. The 
 crisis of her travail comes on the tormented body 
 and mind of the Duchess (II. 1) to the swift 
 accompaniment of Bosola's mockery. Brachi- 
 ano's wooing, and his later recapture, of Vittoria, 
 take on the sick dreadfulness of figures in a 
 nightmare, whose shadows parody them with ob- 
 scene caricature; because of the ceaseless ape- 
 like comments of Flamineo, cold, itchy, filthily 
 knowing. 
 
 Light has interestingly been thrown of late on 
 Webster's method of composition. It had long 
 been known that he repeats a good many lines 
 and phrases from himself and from other peo- 
 ple: and that a great deal of his writing, espe- 
 cially in his best and most careful work, has the 
 air of being proverbial, or excerpt. John Ad- 
 dington Symonds remarked with insight a good 
 many years ago that Webster must have used 
 a note-book. His plays read like it. And now 
 Mr. Crawford has discovered some of the sources 
 he compiled his note-book from.^ 
 
 * Crawford, Collectanea, i. 20-46, ii. 1-63. 
 
146 JOHN WEBSTER 
 
 It would be useless to repeat Mr. Crawford's 
 list with a few additions, or to examine the in- 
 stances one by one. Nearly, not quite, all his 
 cases seem to me to be real ones. There are cer- 
 tainly quite enough to enable one to draw impor- 
 tant inferences about Webster's way of working. 
 These instances of borrowing are very numerous, 
 and chiefly from two books, Sidney's Arcadia, 
 K and Montaigne — favourite sources of Eliza- 
 bethan wisdom. They are very clearly marked, 
 and consist in taking striking thoughts and 
 phrases in the original, occasionally quite long 
 ones, and rewriting them almost verbally, some- 
 times with slight changes to make them roughly 
 metrical. It is a quite different matter from the 
 faint "parallels" of ordinary commentators. I 
 give one of the more striking instances, to illus- 
 trate : 
 
 Arcadia, Bk. II.: 
 
 "But she, as if he had spoken of a small matter 
 when he mentioned her life, to which she had not 
 leisure to attend, desired him, if he loved her, to 
 shew it in finding some way to save Antiphilus. For 
 her, she found the world but a wearisome stage 
 unto her, where she played a part against her will, 
 and therefore besought him not to cast his love 
 in so unfruitful a place as could not love it- 
 self. . . ." 
 
CHARACTERISTICS 147 
 
 Arcadia, Ek. III.: 
 
 "It happened, at that time upon his bed, towards 
 the dawning of the day, he heard one stir in his 
 chamber, by the motion of garments, and with an 
 angry voice asked who was there. *A poor gentle- 
 woman,' answered the party, 'that wish long life 
 unto you.' 'And I soon death unto you,' said he, 
 'for the horrible curse you have given me.' " 
 
 The Duchess of Malfi, IV. 1 (p. 85) : 
 
 Duchess. "^Vho must dispatch me? 
 
 I account this world a tedious theatre 
 
 For I do play a part in't 'gainst my will." 
 BosoLA. "Come, be of comfort; I will save your life.** 
 Duchess. "Indeed, I have not leisure to tend 
 
 So small a business." 
 BosoLA. "Now, by my life, I pity you." 
 Duchess. "Thou art a fool, then. 
 
 To waste thy pity on a thing so wretched 
 
 As cannot pity itself. I am full of daggers. 
 
 Puff, let me blow these vipers from me! 
 
 What are you.^" 
 
 Enter Servant. 
 Servant. "One that wishes you long life." 
 Duchess. "I would thou wert hang'd for the horrible 
 curse 
 
 Thou hast given me." 
 
 There are three explanations of all this. 
 Either Webster knew the Arcadia so well that 
 he had a lot of it by heart. Or he had the book 
 and worked from it. Or he kept a note-book, 
 into which he had entered passages that struck 
 him, and which he used to write the play from. 
 
148 JOHN WEBSTER 
 
 It seems to me certain that the third is the true 
 explanation. We know that Elizabethan authors 
 did sometimes keep note-books in this way. Ba- 
 con did so, and Ben Jonson, whom Webster ad- 
 mired and rather resembled, worked most me- 
 thodically this way. The memory theory could 
 scarcely explain the verbal accuracy of so many 
 passages. But there are other considerations, 
 which make the note-book probable. The pas- 
 sages from the Arcadia or from Montaigne came 
 very often in lumps. You will get none, or only 
 one or two, for some scenes, and then twenty 
 lines or so that are a cento of them, carefully 
 dovetailed and worked together. It is very diffi- 
 cult to imagine a man doing this from memory 
 or from a book. But it is exactly what would 
 happen if he were using a note-book which had 
 several consecutive pages with Arcadia extracts, 
 several more with Montaigne, and so on. The 
 passage I quoted, which brings together an ex- 
 tract from Arcadia, III., and another from Ar- 
 cadia, II., exemplifies this. But there are better 
 instances. The first ten lines of The Duchess of 
 Malfi, IV. 1 (p. 84), contain three continuous 
 more or less verbal thefts from different parts 
 of the Arcadia, the first and third from Book II., 
 the second from Book I. Better still; in II. 1 
 (p. 67) , Bosola has to utter some profound "con- 
 
CHARACTERISTICS 149 
 
 templation," worthy of his malcontent type. 
 Webster could not think of anything at the mo- 
 ment. He generally seems to have had recourse 
 to his note-book when he was gravelled ; for a lot 
 of his borrowed passages make very little sense 
 as they come in, and that of a rather sudden na- 
 ture, in the way that generally betokens an in- 
 terrupted train of thought. He went to his note- 
 books on this occasion. He found, probably con- 
 tiguous there, several sentences of a weighty, dis- 
 connected sense. They are from Montaigne, 
 Florio's translation, pages 246, 249, 248, in that 
 order.^ Put together they have, as a matter of 
 fact, very little meaning. 
 
 BosoLA. "O, Sir, the opinion of wisdom is a foul 
 tetter that runs all over a man's body; if sim- 
 plicity direct us to have no evil it directs us to 
 a happy being; for the subtlest folly proceeds 
 from the subtlest wisdom ; let me be simply hon- 
 est." 
 
 Still, it did. And being at his Montaigne note- 
 books, Webster went on. Bosola's next speech 
 but one borrows from the first Book. For the 
 long speech that follows it, he goes back to Book 
 II.; and makes it entirely from two different 
 passages, one on p. 239, one on p. 299. 
 
 A last instance is still more convincing. It 
 
 * Professor Henry Morley's reprint. 
 
150 JOHN WEBSTER 
 
 concerns A Monumental Column, lines 23-35, 
 and The Duchess of Malfi, III. 2 (p. 79), the 
 description of Antonio. The first passage is 
 mostly taken verbally from the two sources, Ben 
 Jonson's Dedication to A Masque of Queens and 
 the description of Musidorus in Arcadia, Book I. 
 The passage in the play contains one of the 
 same lines from Jonson, together with a different 
 part of the sentence describing Musidorus, and 
 a couple of lines from another part of Arcadia, 
 Book I. And the remainder of the description 
 of Musidorus duly turns up in The Duchess of 
 Malfi a few scenes later, in IV. 1 (p. 84), sand- 
 wiched between two passages from Arcadia, 
 Book II. 
 
 A good many of these passages Webster cop- 
 ied out identically, except sometimes for a few 
 changes to make them go into rough verse. 
 Others he altered in very interesting ways. It 
 was not necessarily part of his goodness as an 
 author to alter them. His genius comes out 
 equally in the phrases he used to produce far 
 greater eff*ect than they do in the original, by 
 putting them at some exactly suitable climax. 
 We are getting beyond the attitude, born of the 
 industrial age and the childish enthusiasm for 
 property as such, which condemns such plagiar- 
 ism, imitation, and borrowing. The Elizabethans 
 
CHARACTERISTICS 151 
 
 had for the most part healthy and sensible views 
 on the subject. They practised and encouraged 
 the habit. When Langbaine, in his preface to 
 Momus Triumphans, "condemns Plagiaries" 
 (though he is only thinking of plots, even then), 
 it is a sign of the decadence towards stupidity. 
 The poet and the dramatist work with words, 
 ideas, and phrases. It is ridiculous, and shows 
 a wild incomprehension of the principles of lit- 
 erature, to demand that each should only use his 
 own ; every man's brain is filled by thoughts and 
 words of other people's. Webster wanted to 
 make Bosola say fine things. He had many 
 in his mind or his note-book : some were borrowed, 
 some his own. He put them down, and they an- 
 swer their purpose splendidly. 
 
 / "I stand like one 
 
 l/^ That long hath ta'en a sweet and golden dream; 
 I am angry with myself, now that I wake/' 
 
 That was, or may have been, of his own inven- 
 tion. 
 
 "The weakest arm is strong enough that strikes 
 With the sword of justice." 
 
 That he had found in Sidney. There is no dif- 
 ference. In any case the first, original, passage 
 was probably in part due to his friends' influ- 
 ence; and the words he used were originally 
 
152 JOHN WEBSTER 
 
 wholly "plagiarised" from his mother or his 
 nurse-maid. "Originality" is only plagiarising 
 from a great many. 
 
 So Webster reset other people's jewels and 
 redoubled their lustre. "The soul must be held 
 fast with one's teeth . . ." he found Mon- 
 taigne remarkably saying in a stoical passage. 
 The phrase stuck. Bosola, on the point of death, 
 cries : ^ 
 
 "Yes I hold my weary soul in my teeth; 
 'Tis ready to part from me." 
 
 It is unforgettable. 
 
 Webster improved even Donne, in this way; 
 in a passage of amazing, quiet, hopeless pathos, 
 the parting of Antonio and the Duchess {Duch- 
 ess of Malfi, III. 5), which is one long series of 
 triumphant borrowings: 
 
 *'We seem ambitious God's whole work to undo; 
 Of nothing He made us, and we strive too 
 To bring ourselves to nothing back," 
 
 Donne writes in An Anatomy of the World. 
 
 "Heaven fashion'd us of nothing; and we strive 
 To bring ourselves to nothing," 
 
 are Antonio's moving words. 
 
 * It is only because there are scores of other certain borrowings 
 of Webster from Montaigne that I accept this one. By itself it 
 would not be a convincing plagiarism. 
 
CHARACTERISTICS 153 
 
 This last example illustrates one kind of the 
 changes other than metrical Webster used to 
 make. He generally altered a word or two, 
 with an extraordinarily sure touch, which proves 
 his genius for literature. He gave the passages 
 life and vigour, always harmonious with his own 
 style. You see, by this chance side-light, the 
 poet at work, with great vividness. "Fashion'd" 
 for "made" here, is not a great improvement; 
 but it brings the sentence curiously into the key 
 of the rest of the scene. The metrical skill is 
 astounding — the calm weight of "fashion'd"; the 
 slight tremble of "Heaven" at the beginning of 
 the line; the adaptation from Donne's stiff heavy 
 combative accent, the line ending with "and we 
 strive too," to the simpler easier cadence more 
 suited to speech and to pathos, "... ; and we 
 strive"; and the repetition of "nothing" in the 
 same place in the two lines. 
 
 The long first example I gave of borrowing 
 from Sidney gives good instances of change, 
 among others the half-slangy vividness of 
 
 "Thou art a fool, then. 
 To waste thy pity on a thing so wretched 
 As cannot pity itself ...,'* 
 
 for Sidney's mannered, dim, 
 
 "and therefore besought him not to cast his 
 love in so unfruitful a place as could not love itself." 
 
154 JOHN WEBSTER 
 
 But the same places in The Duchess of Malfi 
 and the Arcadia have a much finer example. The 
 description of Queen Erona is transferred to the 
 Duchess again. Sidney says that in her soitow, 
 one could "perceive the shape of loveliness more 
 perfectly in woe than in joyfulness." Webster 
 turned this, with a touch, to poetry in its sheerest 
 beauty. 
 
 BosoLA. "You may discern the shape of loveliness 
 
 More perfect in her tears than in her smiles." 
 
 It is just this substitution of the concrete for 
 the abstract — which is the nearest one could get 
 to a definition of the difference between a thought 
 in good prose and the same thought in good 
 poetry — that Webster excels in. Even where his 
 adjectives gain, it is in this direction. 
 
 "Or is it true that thou wert never but a vain 
 name, and no essential thing?" 
 
 says Sidney in a long passage on Virtue. Web- 
 ster makes it a shade more visual, and twenty 
 times as impressive: 
 
 "Or is it true thou art but a bare name. 
 And no essential thing.'*" 
 
 So Bosola gives life to a meditation of Mon- 
 taigne. Montaigne's democratic mind pondered 
 
CHARACTERISTICS 155 
 
 in his study on the essential equahty of men. 
 "We are deceived," he says of princes; "they are 
 moved, stirred, and removed in their motions by 
 the same springs and wards that we are in ours. 
 The same reason that makes us chide and brawl 
 and fall out with any of our neighbours, causeth 
 a war to follow between princes ; the same reason 
 that makes us whip or beat a lackey maketh a 
 prince (if he apprehend it) to spoil and waste a 
 whole province. . . ." Bosola is the heart of 
 democracy. "They are deceived, there's the same 
 hand to them; the like passions sway them; the 
 same reason that makes a vicar to go to law for 
 a tithe-pig, and undo his neighbours, makes them 
 spoil a whole province, and batter down goodly 
 cities with the cannon." The tithe-pig carries 
 you on to Parnassus; Bosola has the vision of an 
 artist. 
 
 The liveliness of the "there's" for "there is" 
 in the last quotation is typical. Webster, like all 
 the great Elizabethans, knew he was writing for 
 the ear and not the eye. They kept in close 
 touch, in their phrases, rhythms, and turns, with 
 speech. Their language was greater than speech, 
 but it was in that kind; it was not literature. 
 
 But there is one example of adoption and 
 adaptation where Webster stands out quite clear 
 as the poet, with the queer and little-known men- 
 
156 JOHN WEBSTER 
 
 tal processes of that kind of man suddenly 
 brought to the hght. Montaigne has a passage : 
 
 "Forasmuch as our sight, being altered, repre- 
 sents unto itself things alike ; and we imagine that 
 things fail it as it doth to them: As they who 
 travel by sea, to whom mountains, fields, towns, 
 heaven, and earth, seem to go the same motion, 
 and keep the same course they do." 
 
 The sense is clear and on the surface. He is 
 illustrating the general rule by an interesting 
 instance from ordinary experience. When you 
 go in a train, or a boat, the sky, the earth, and 
 its various features, all seem to be moving in one 
 direction.^ In The White Devil Flamineo is 
 tempting Vittoria with the happiness Brachiano 
 can give her. 
 
 "So perfect shall be thy happiness, that, as men 
 at sea think land and trees and ships go that way 
 they go, so both heaven and earth shall seem to go 
 your voyage/' 
 
 Webster took this instance of Montaigne's and 
 used it to help out quite a different sense. He 
 used it as a simile of that elusive, unobvious, im- 
 aginative kind that illuminates the more that you 
 can scarcely grasp the point of comparison. But 
 
 * Note, though, that Montaigne has made a slip. They really 
 appear to be moving in the opposite direction to yourself. Web- 
 ster takes the idea over, mistake and all. 
 
CHARACTERISTICS 157 
 
 he did more. He was led to it by thinking, as a 
 poet thinks, only half in ideas and half in words. 
 Or rather, with ordinary people, ideas lead to 
 one another, suggest one another, through ideas. 
 With poets they do it through words, quite illogi- 
 cally. The paths of association in the brain are 
 different in the two cases. A word is an idea 
 with an atmosphere, a hard core with a fringe 
 round it, like an oyster with a beard, or Profes- 
 sor William James' conception of a state of 
 mind. Poets think of the fringes, other people 
 of the core only. More definitely, if the diction- 
 ary meaning of a word is a and the atmosphere 
 oc, the poet thinks of it as ( oo + a)^ and his trains 
 of thought are apt to go on accordingly. So 
 here, Webster found, vaguely, "heaven and 
 earth' ' . . . "going the same motion" . . . and 
 he leapt to the mystical conception of supreme 
 happiness. He took "heaven and earth" from 
 their original, half material, significance, and 
 transfigured them. He took them from the illus- 
 tration and put them into the thing illus- 
 trated. The meaning of the original suggested 
 one thing to his mind, the words another; he 
 combined them, in another world. And the re- 
 sult is a simile of incomprehensible appropriate- 
 ness and exquisite beauty, an idea in a Shelleyan 
 altitude where words have various radiance rather 
 
158 JOHN WEBSTER 
 
 than meaning, an amazing description of the 
 sheer summit of the ecstasy of joy. 
 
 The note-book habit suited those idiosyncrasies 
 of Webster's slow-moving mind which distin- 
 guished him from the ready rhetoric of Fletcher 
 and the perpetual inspiration of Shakespeare. 
 The use of such a thing by a poet implies a dif- 
 ference from other poets in psychology, not, as 
 is often ignorantly supposed, in degree of merit. 
 It merely means he has a worse memory. All 
 writers are continually noting or inventing 
 phrases and ideas, which form the stuff from 
 which their later inspiration chooses. Some have 
 to note them down, else they slip away for ever. 
 Others can note them in their mind and yet feel 
 secure of retaining them. The advantage of this 
 method is that you unconsciously transmute all 
 "borrowed" ideas to harmony with your own per- 
 sonality — that when you hunt them out to re- 
 claim them you find them slightly changed. The 
 disadvantage, under modern conditions, is that 
 you may conmiit the most terrible sin of plagiar- 
 ism, and lift another man's work, and display it 
 in a recognisable form, without knowing it. So 
 Meredith in one of his last and best lyrics, an 
 eight-lined poem called "Youth and Age," re- 
 peats a line identically from Swinburne's best 
 poem, The Triumph of Time; and all uncon- 
 
CHARACTERISTICS 159 
 
 sciously. The disadvantage of the note-book 
 method is that you have to perform the operation 
 of digesting your trophy, harmonising it with the 
 rest of the work, on the spot. Webster does not 
 always do this successfully. There are passages, 
 as we have seen, where he too flagrantly helps 
 himself along with his note-book. But as a rule 
 he weaves in his quotations extraordinarily well; 
 they become part of the texture of the play, 
 adding richness of hue and strength of fabric. 
 In The White Devil, in the scene of astounding 
 tragical farce where Flamineo persuades Vittoria 
 and Zanche to try to murder him with bulletles« 
 pistols, the quotations from Montaigne come in 
 entirely pat. For it is not, generally, when the 
 play goes slowest that Webster has most recourse 
 to his note-book. The swift passion of Ferdi- 
 nand's interview with the guilty Duchess {Duch- 
 ess of Malfi, III. 2) is, if you enquire closely, 
 entirely composed of slightly altered passages 
 from the Arcadia, This detracts no whit from 
 its tumultuous force. 
 
 The chief value of working through a note- 
 book, from a literary point of view% is this. A 
 man tends to collect quotations, phrases, and 
 ideas, that particularly appeal to and fit in with 
 his own personality. If that personality is a 
 strong one, and the point of his work is the 
 
160 JOHN WEBSTER 
 
 pungency with which it is imbued with this strong 
 taste, the not too injudicious agglutination of 
 these external fragments will vastly enrich and 
 heighten the total effect. And this is, on the 
 whole, what happens with Webster. The heap- 
 ing-up of images and phrases helps to confuse 
 and impress the hearer, and gives body to a taste 
 that might otherwise have been too thin to carry. 
 Webster, in fine, belongs to the caddis-worm 
 school of writers, who do not become their com- 
 plete selves until they are incrusted with a thou- 
 sand orts and chips and fragments from the 
 world around. 
 
 It would be possible to go on for a long time 
 classifying various characteristics of Webster, 
 and discovering them in different passages or 
 incidents in his plays. And it would be possible, 
 too, to lay one's finger on several natural reac- 
 tions and permanent associations in that brain. 
 All have noticed his continual brooding over 
 death. He was, more particularly, obsessed by 
 the idea of the violence of the moment of death. 
 Soul and body appeared to him so interlaced that 
 he could not conceive of their separation without 
 a struggle and pain. Again, his mind was al- 
 ways turning to metaphors of storms and bad 
 weather, and especially the phenomenon of light- 
 
CHARACTERISTICS 161 
 
 ning. He is for ever speaking of men lightening 
 to speech or action; he saw words as the flash 
 from the thunder-cloud of wrath or passion. 
 
 But, after all, the chief characteristic of Web- 
 ster's two plays and of many things in those 
 plays, is that they are good; and the chief char- 
 acteristic of Webster is that he is a good drama- 
 tist. The great thing about The Duchess of 
 Malfi is that it is the material for a superb play; 
 the great thing about the fine or noble things in 
 it is not that they illustrate anything or belong 
 to any class, but, in each case, the fine and noble 
 thing itself. All one could do would be to print 
 them out at length ; and this is no place for that ; 
 it is easier to buy Webster's AVorks (though, in 
 this scandalous country, not very easy). The 
 end of the matter is that Webster was a great 
 writer; and the way in which one uses great 
 writers is two-fold. There is the exhilarating 
 way of reading their writing; and there is the 
 essence of the whole man, or of the man's whole 
 work, which you carry away and permanently 
 keep with you. This essence generally presents 
 itself more or less in the form of a view of the 
 universe, recognisable rather by its emotional 
 than by its logical content. The world called 
 Webster is a peculiar one. It is inhabited by 
 people driven, like animals, and perhaps like 
 
162 JOHN WEBSTER 
 
 men, only by their instincts, but more blindly 
 and ruinously. Life there seems to flow into its 
 forms and shapes with an irregular abnormal and 
 horrible volume. That is ultimately the most 
 sickly, distressing feature of Webster's charac- 
 ters, their foul and indestructibLgjotallty. It fills 
 one with tEeTepiHsiorrone feels at the unending 
 soulless energy that heaves and pulses through 
 the lowest forms of life. They kill, love, torture 
 one another blindly and without ceasing. A 
 play of Webster's is full of the feverish and 
 ghastly turmoil of a nest of maggots. Maggots 
 are what the inhabitants of this universe most 
 suggest and resemble. The sight of their fever 
 is only alleviated by the permanent calm, un- 
 friendly summits and darknesses of the back- 
 ground of death and doom. For that is equally 
 a part of Webster's universe. Human beings 
 are writhing grubs in an immense night. And 
 the night is without stars or moon. But it has 
 sometimes a certain quietude in its darkness; 
 but not very much. 
 
APPENDICES. 
 
Appendix A. — "Appius and Virginia" 
 
 [The original form of this appendix was rearranged 
 and shortened by the author for separate publication 
 in the Modern Languages Remew, vol. viii. No. 4 (Octo- 
 ber, 1913). I have here combined the two versions, fol- 
 lowing the order of the second, but restoring most of 
 the passages which were omitted from it to save space. 
 
 E. M.] 
 
THE AUTHORSHIP OF THE LATER 
 "APPIUS AND VIRGINIA." ' 
 
 It is startlingly obvious, and has been remarked by 
 every critic of Webster, that Appius and Virginia is 
 quite different from his other plays. It "stands apart 
 from the other plays," says Professor Vaughan.^ Dr. 
 Ward recognises it as a work of Webster's "later man- 
 hood, if not of his old age." Mr. Wilham Archer vastly 
 prefers it to the ordinary crude Websterian melodrama. 
 In fact, critics, whether of the Ehzabethans in general 
 or of Webster in particular, have always exhibited 
 either conscious discomfort or unconscious haste and 
 lack of interest, when they came to this play. As they 
 have never questioned its authenticity, their perfunc- 
 tory and unprofitable treatment of it is noteworthy. 
 They cannot fit it in. In summing up Webster's charac- 
 teristics, they have either quietly let it slide out of sight, 
 or else brought it formally and unhelpfully in, to sit 
 awkward and silent among the rest like a deaf unpleas- 
 ant aunt at a party of the other side of the family. But 
 never, so far as I am aware, has anyone suggested that 
 it is not by Webster. 
 
 We may sympathise with the critics. The more 
 closely Appius and Virginia is looked at, the less it 
 shows of the Webster we know. With Northward Ho 
 
 *The only other Appius and Virginia known is the old-fashioned 
 lumbering play by "R. B." (probably Richard Bower) of 1576. 
 
 a C. H. E. L., vol. vi, p. 182. 
 
 165 
 
166 JOHN WEBSTER 
 
 and Westward Ho, one is not discomforted at finding 
 almost no such mark. You may imagine Webster a 
 young man, collaborating with an older, in a well- 
 defined, not very congenial, type of play, contributing 
 the smaller part. There are a hundred reasons against 
 what we mean by Webster being prominent in those 
 plays. Anyhow, a young man's work is frequently any- 
 body's ; especially his hack-work. Who could pick out 
 Meredith's war correspondence from anyone else's ? But 
 once he has developed his particular savour, it can 
 hardly fade into commonness again. It is as with faces. 
 You can often mistake two young faces. But once the 
 soul has got to work, wrinkling and individualising the 
 countenance, it remains itself for ever, even after the 
 soul has gone. The taste we recognise as Webster de- 
 veloped between 1607 and 1615. It is a clinging, un- 
 mistakable one. Later on he imitated models who pro- 
 voked it less powerfully. But a close, long scrutiny, 
 before which Appius and Virginia grows more cold and 
 strange, increasingly reveals Webster in The Devil's 
 Law-Case, even in A Cure for a Cuckold, of which he 
 only wrote part. 
 
 Examine Appius and Virginia sesthetically and as 
 a whole. Webster is a dogged, slow writer, and roman- 
 tic — in the sense that single scenes, passages, or lines 
 have merit and intensity on their own account. As a 
 rule, he finely proves that quintessence of the faith that 
 the God of Romanticism revealed to his inattentive 
 prophet. "Load every rift with ore." And there is a 
 kind of dusty heat over all. Appius and Virginia is 
 precisely the opposite. Its impression is simple and 
 cool. It seems more an effort at classicism — uncon- 
 
APPENDICES 167 
 
 scious perhaps. There are not many lines or images 
 you stop over. You see right to the end of the road. 
 It is, of course, a very poor argument against at- 
 tributing a play to any particular author, that he has 
 not written this kind of play elsewhere. The very fact 
 that he hasn't, makes it all the harder to know what 
 his attempt in this manner would be like. And when 
 such an argument is used, as it is, to prove that A 
 Yorkshire Tragedy is not Shakespeare's, it is of no 
 value, though it may be on the right side. What is 
 permissible, however, is, when a writer has several dis- 
 tinct characteristics, to expect to recognise some of 
 them, when he is seriously attempting a kind of play 
 not very different from his ordinary one ; especially if 
 these characteristics are of certain kinds. A mere jour- 
 nalist, turning out his daily task, may sometimes write 
 an indistinguishable undistinguished play in a different 
 style. A great master of a certain type may possibly^ 
 his tongue just perceptibly bulging the cheek, flash out 
 something quite good in an entirely other kind, as a tour 
 de force. Or a very brilliant and not at all serious per- 
 son, with a trick of writing, some Grceculus of literature, 
 may sink his own personality entirely in the manner of 
 another. But that is only possible if he is able to aim 
 entirely at parody, and not at all at art. Few artists 
 could ever do this. In any case, Webster and Appms 
 and Virginia do not fit into any of these potential ex- 
 planations. He worked (as he tells us, and we can see) 
 slowly and with trouble. Both his method and the result 
 show that he was no easily adaptable writer. His 
 clumsy, individual, passionate form betrays itself under 
 borrowed clothes. This does not mean that he strode 
 
168 JOHN WEBSTER 
 
 always intensely and unswervingly along his own path. 
 He was, in an odd way, ready enough to put on other 
 people's clothes that did not suit him. But they. never 
 fitted all over. It is suggested that in Appius and Vir- 
 ginia he was trying to imitate Shakespeare's Roman 
 tragedies. This might explain the absence of some of 
 his peculiarities, and the presence of other marks ; the 
 change of atmosphere, the greater number of rhyming 
 lines, and so forth. But subtler questions of metre and 
 vocabulary go deeper, in proportion as they are more 
 unconscious. Consideration of such delicate points, to- 
 gether with a careful general aesthetic tasting of the 
 whole play, seem to me to warrant a very strong critical 
 doubt whether Webster wrote Appius and Virginia. 
 
 The characters of the play are slight and ordinary. 
 The clown is quite unlike anything we could expect Web- 
 ster to invent. Appius, the Machiavellian villain, has a 
 little fire. Virginius is a mere stage-creature, and, as 
 that, quite creditable. Virginia is a virgin. The crowd 
 of soldiers is a soldiers' crowd. Webster's characters, 
 in other plays, if they do not always (compared at least 
 with Shakespeare's) make a highly individual impres- 
 sion on the mind, always leave a dent. 
 
 The metre of Appius and Virginia is not Webster's. 
 The blank verse is much stricter. Webster's loose, im- 
 pressionistic iambics, with their vague equivalence and 
 generous handling, are very unlike these regular, rhe- 
 torical lines. Webster's great characteristic of begin- 
 ning a line with what classical prosodists would call 
 an anapaest finds no place here. And the general metri- 
 cal technique of which this is only the most obvious 
 manifestation — the continual use of substitution and 
 
APPENDICES 169 
 
 equivalence in the feet, or, better, the thinking more in 
 lines and less in feet ^ — is strikingly absent in Appms 
 and Virginia. These prosodic habits are also almost as 
 little prominent in the possibly Websterian part of 
 A Cure for a Cuckold. But there is another point 
 which marks Appius and Virginia off from all the rest. 
 In the other plays, there is little attempt to keep a line 
 that is divided between two speakers pentametrical. If 
 one speech ends with a line of two and a half feet, the 
 next may begin with a line of two feet, or of three, or 
 with a complete line. Appius and Virginia keeps al- 
 most invariably to the old tradition, by which the 
 speeches dovetail perfectly.^ 
 
 The first and almost the only characteristic in this 
 play to strike a casual reader, is the vocabulary. It 
 is full of rare Latin words, mostly wearing an air of 
 recent manufacture; "to deject" (in a literal sense), 
 "munition," "invasive," "devolved," "donative," 
 "palped," "enthronised," "torved," "strage," and 
 many more. This particular vocabulary is a mark of 
 certain writers, especially of the period at the end of 
 the sixteenth and beginning of the seventeenth cen- 
 
 ^E. g. Duchess of Malfi, III. 2: 
 
 "Did you ever in your life know an ill painter 
 Desire to have his dwelling next door to the shop 
 Of an excellent picture maker?" 
 ' For the perplexing metrical part which Apj)ius and Virginia 
 plays, see the metrical table on p. 190 of Dr. Stoll's John Webster. 
 Its resemblance to A Cure for a Cuckold is only in some direc- 
 tions, and more statistical than real. The metre of both is rather 
 smooth; but in a very different way. It is, of course, rather 
 risky to lay much emphasis on A Cure for a Cuckold: it may have 
 been worked over by Rowley. 
 
170 JOHN WEBSTER 
 
 turies, which had a joyous fertility in inventing new 
 words that soon drooped and grew sterile. It was 
 mostly employed by the slightly classicist authors. Of 
 the major dramatists, Ben Jonson had a touch of it; 
 Marston, Heywood, Chapman, and Shakespeare show 
 it chiefly. Shakespeare has this variety among all his 
 other varieties, neologisms, and airai Xeyofxeva: Chap- 
 man and Heywood this in especial. 
 
 In this and every notable respect the language of 
 Appius and Virginia is unlike Webster's. Whatever 
 linguistic point of detail you choose, the lack of re- 
 semblance is obvious. To take one instance: Dr. Stoll 
 (p. 40), in trying to prove the Webster authorship of 
 the major part of A Cure for a Cuckoldy uses as a test 
 the occurrence of the exclamation "Ha !" especially as 
 comprehending a whole speech. He says it is unusually 
 frequent in Webster. "It appears in The White Devil 
 thirteen times, six of them being whole speeches; in 
 Malfi ten times, two of them whole speeches; in the 
 Law-Case nine times, four of them whole speeches; in 
 Appius and Virginia twice ; in the main plot of the Cure 
 for a Cuckold seven times, two of them whole speeches." 
 The oddness of the Appius and Virginia figures does 
 not strike Dr. Stoll, who is on other business. He ex- 
 plains them, vaguely, by "the frigidity and academic 
 character of the play"; which is far from fair to the 
 slightly Marlovian and "Machiavellian" nature of much 
 of Appius and Virginia. It is not a Jonsonian Roman 
 play. There is no reason why Appius should not have 
 
APPENDICES 171 
 
 said "Ha!" thirteen times, six of tliem whole speeches, 
 except that the author did not write like that. 
 
 Again, the word "foul" was, characteristically, a 
 common one with Webster. It occurs often in The 
 White Devil, on almost every page in The Duchess of 
 Malfi. "Think on your cause," says Contarino to 
 Ercole in The DeviVs Law-Case, II. 2 ; "It is a won- 
 drous foul one." And when the real "devil's law-case" 
 comes on (IV. S), the shameless Winifred desires, 
 "Question me in Latin, for the cause is very foul." 
 There was this habit in Webster of thinking of such 
 moral rottenness as "foul," slightly materialising it. A 
 reader would feel safe in betting that Webster would 
 use the word several times in connection with the trial 
 of Virginia. One knows his comment on it, as one 
 knows how a friend will take a piece of news. The 
 word does not occur in this passage. 
 
 Analysis might find a thousand more points, positive 
 and negative, in which the style and vocabulary of 
 Appius and Virginia are obviously not those of Web- 
 ster. The dissimilarity becomes still more obvious 
 when the language is unanalytically tasted as a whole. 
 It is throughout rhetorical and easy, with a slight 
 permanent artificiality. The style is rather imitative 
 of Shakespeare's, and alive, but not kicking. 
 
 In the general construction and handling of the 
 play there is an un-Websterian childishness and crud- 
 ity. Webster could be gauche enough at times, but 
 not in this shallow, easy way. I need only enumerate 
 some of the instances. 
 
172 JOHN WEBSTER 
 
 The Elizabethans were splendidly unsubservient to 
 time. But the better dramatists tended to conceal their 
 freedom; Webster among them. The keenest-witted 
 spectator of A Midsummer Niglifs Dream or The Mer- 
 chant of Venice could not, unless he were looking for 
 them, discern the tricks Shakespeare has played with 
 time. The instance in Appius and Virginia is far more 
 flagrant, though it might strike an Elizabethan less 
 than us. Act V. scene 3 takes place in the prison. 
 Icilius, seeing Virginius relent towards Appius, vanishes 
 to fetch the body of Virginia. Seven lines after his 
 exit, a shout is heard. It turns out that in this time 
 Icilius has gone through the streets to where Virginia 
 is lying, taken up the body, and started back through 
 the streets carrying it; and the people have begun to 
 make an uproar. Eleven lines later, Icilius enters with 
 the body. If the play stands as it was written, it is 
 difficult to believe that Webster could have committed 
 such absurdities. They might possibly, but not prob- 
 ably, be explained by a theory, for which there is other 
 evidence, that we have the play in a cut and revised 
 state.* But nothing can be thought too childish to 
 come from the author of the crowd-effects in Act II. 2, 
 where the First Soldier asks : 
 
 Soldiers, shall I relate the grievances 
 Of the whole regiment? 
 
 You might expect Omnes to answer "Yes !" or "No !" 
 if they were all agreed. It is too startling when, with 
 
 ^ See page 200. 
 
APPENDICES 173 
 
 one voice, they cry "Boldly !" But a more amazing in- 
 stance of sympathy and intelligence follows. The First 
 Soldier ends a piece of rhetoric with: 
 
 from thence arise 
 A plague to choke all Rome ! 
 Omnes. And all the suburbs ! 
 
 There is a childishness that goes deeper, in the hand- 
 ling of the plot and episodes. It is all told with a 
 forthright and unthinking simplicity that is quite dif- 
 ferent from any Chapmanesque stark directness; the 
 simplicity of a child who wants to tell a story, not of 
 an artist who grasps the whole. It is apparent in the 
 soliloquies of II. 1, in the end of I. 3, and especially at 
 the beginning of the same scene, in the interview be- 
 tween Marcus and Appius. Appius is melancholy, 
 declares himself in love. Marcus asks with whom, offer- 
 ing to act pander. Appius tells him, Virginia. 
 
 Marcus. Virginia's ! 
 
 Appius. Hers. 
 
 Marcus. I have already found 
 
 An easy path which you may safely tread. 
 
 Yet no man trace you. 
 
 He goes on to explain in detail his rather elaborate 
 plan. 
 
 It is difficult to imagine dramatic innocence of this 
 kind coming from Webster, whose humour and bizarrerie 
 are, if not always successful, always entirely conscious, 
 and whose simplicity, as playwright, is rather archais- 
 tic than childish. 
 
174 JOHN WEBSTER 
 
 These are some of the immediate difficulties in be- 
 lieving Appius and Virginia to be by Webster. The 
 further difficulties of explaining the nature and date 
 of the play, if it is by him, strengthen our incredulity. 
 How Webster came to write such a play, his various 
 critics and commentators have not tried to ex- 
 plain; chiefly because they have not understood that 
 there was any need of explanation. They have realised 
 neither how astonishing a tour de force it is, for an 
 author so completely to sink his personality, nor that 
 Webster is the last man to be capable of such a feat. 
 The dumb evidence of their inability to make this play 
 fit in with or illuminate the rest of Webster's work, 
 speaks for them. When Webster wrote it, is a ques- 
 tion they have tried to answer, however dimly. Their 
 answers have all been different, and all importantly un- 
 convincing. In the first place, the whole style of the 
 play, in plot, characterisation, and metre, suggests an 
 early date, somewhere between 1595 and 1615; and 
 joins it, loosely, with Julius Ccesar (1601?), Coriolanus 
 (1608?) and Heywood's The Rape of Lucrece (1604?). 
 This is especially to be remarked of the metre, which 
 is rather formal, without being stiff. It has few 
 "equivalences," that is to say, the lines have nearly 
 always ten (or, if "feminine," eleven) syllables. The 
 licences are regular. They mostly consist of a few 
 limited cases in which elision occurs, always noticeably, 
 and almost conventionally — the chief example is be- 
 tween "to" and a verb beginning with a vowel.^ I have 
 * E. g. "To obey, my lord, and to know how to rule . . ." 
 
APPENDICES 175 
 
 already noticed the metrical dovetailing of speeches. 
 All these prosodic characteristics suit, some rather 
 demand, a date between 1600 and 1610. So does the 
 influence of Marlowe and Machiavellism, and the char- 
 acter of the clown, Corbulo, who is staringly introduced 
 into the original story. Finally, the general and spe- 
 cific dissimilarity in style of Appius and Virginia and 
 Webster's other plays forbids a middle date, and re- 
 quires an early rather than a late one, if the play be 
 his. Only a young hand could have disguised its indi- 
 viduality so completely. 
 
 The other evidence, however, points in precisely the 
 opposite direction. When you try to suggest a possi- 
 ble date you meet bewildering difficulties. One of the 
 most certain things about Appius and Virginia is that 
 it is strongly influenced by Shakespeare's Roman plaj^s, 
 and especially by Coriolanus} Coriolanus is dated by 
 most critical opinion as 1608-9. So Appius and Vir- 
 gvnia must be at least as late as 1609. But that is 
 definitely in Webster's middle, most individual, period. 
 The White Devil appeared in 1611, and he was con- 
 fessedly a long time in writing it. If the author of The 
 White Devil wrote Appius and Virginia, it cannot have 
 been only a year or eighteen months before. Then 
 again you cannot slip the Roman play amazingly be- 
 tween The White Devil and The Duchess of Malfi (c. 
 1613). It would be far easier to say that Shakespeare 
 wrote Titus Andronicus between As You Like It and 
 
 * StoU, pp. 193-197, illustrates this fully enough. A single read- 
 ing of the play will prove it. 
 
176 JOHN WEBSTER 
 
 Twelfth Night. And you must leave a decent interval 
 after The Duchess of Malfi. You feel inclined to drop 
 it quietly in the vacant space between The Duchess of 
 Malfi and The DeviVs Law-Case. But the progression 
 in style here is so clear and gradual that it is nearly 
 as difficult to squeeze it in there as between the trage- 
 dies. Besides, if you get as late as 1617 or 1618, you 
 may as well listen to Dr. Stoll's evidence — that it is 
 not mentioned in Webster's dedication to The DeviVs 
 Law-Case (printed 1623), and that it shows such close 
 debts to Shakespeare that Webster must have written 
 if after reading the First Folio (1623). So, bufFeted 
 and confused, you take refuge in his spacious "1623- 
 1639"; a date which is in direct opposition to all your 
 first conclusions. And if you want to adorn the affair, 
 now you have settled it, with the circumstance and 
 charm of reality, you may attribute, with Dr. Stoll, 
 not only Webster's style and handling to his study of 
 the First Folio, but his Marlowe characteristics to his 
 recent study of The Massacre at Paris (1593) pre- 
 paratory to writing his own play The Guise, his clown 
 to his friendship with Heywood, his strange style to his 
 imitativeness of the fashion of his time, and his writing 
 this sort of play at all to his fancy for going back to 
 the fashions of twenty or thirty years earlier! 
 
 II 
 
 Well then, what reasons are there for thinking that 
 Webster did write Appius and Virginia? The reasons 
 
APPENDICES 177 
 
 are two — the attribution in 1654, and repetitions or 
 parallels between Webster's other plays and this. They 
 require examination. 
 
 Appius and Virginia was printed and published in 
 1654, as by John Webster. The same edition was put 
 forth in 1659 with a new title-page "Printed for Hum- 
 phrey Moseley";^ and again in 1679, "Acted at the 
 Duke's Theatre under the name of The Roman Virgin 
 or Unjust Judge.'' It is possible that Moseley only 
 took over the edition between 1654 and 1659. In that 
 case the attribution has even less weight. But let 
 us put it at its strongest and suppose (what is most 
 probable) that Moseley was always the publisher. It 
 is being realised more and more how little importance 
 attributions of the second half of the seventeenth cen- 
 tury have. The theatrical traditions had been broken. 
 Publishers attributed by guess-work, or hearsay, or to 
 sell the book. In 1661, Kirkman pubHshed The Thra- 
 cian Wonder as by Webster and Rowley. "No one," 
 says Professor Vaughan, "except the editor, has ever 
 supposed that Webster can have had a hand in it." 
 Yet it is as Websterian as Appius and Virginia. The 
 truth is, critics have at the back of their minds an 
 idea that good poets write good poetry, and bad poets 
 write bad poetry. Since this is as far as they can get, 
 they are ready to give any good poem or play to any 
 
 * For Moseley and his activities, v. Dictionary of National Biog- 
 raphy; Plomer, Dictionary of Booksellers and Printers, 1641-1667; 
 Masson, Life of Milton, iii. 448-457, vi. 352; Parrott, Tragedies 
 of Chapman, p. 683; Malone, Variorvm Shakespeare, iii. 229. 
 
178 JOHN WEBSTER 
 
 good poet, and to refuse any bad one. Appius and 
 Virginia being a fairly good play, there is no reason 
 in the world why it should not be the work of Webster, 
 who was a good writer. The Thraciam Wonder, a bad 
 play, could not possibly be from that hand. . . . The 
 truth is very different. In actuality, a good poet or 
 playwright tends to write good and bad things in his 
 own style. An examination of the works of poets we 
 can be sure about — Keats, or Shelley, or Swinburne — 
 shows this. The author of the sonnet On first looking 
 into Chapman's Homer and the Ode to a Nighti/ngale 
 also wrote the sonnets To my Brother George and to 
 G.A.W. If the work of a century ago were largely 
 anonymous or doubtful, and if the principles of Eliza- 
 bethan criticism were applied, he might be given Alastor 
 or The Vision of Judgement; he would certainly be 
 robbed of the sonnets to George Keats and Georgiana 
 Wylie. 
 
 Humphrey Moseley was, as a matter of fact, one 
 of the more trustworthy publishers of the time. Ma- 
 lone and Professor Parrott are too hard on him. But 
 he had the faults and ignorance of his period. Among 
 other attributions he gives The Merry Devil of Edmon- 
 ton to Shakespeare, The Parliament of Love (Massin- 
 ger) to Rowley, The Faithful Friends to Beaumont and 
 Fletcher, Alphonsus, Emperor of Germany to Chap- 
 man, The Widow (Middleton) to Jonson, Fletcher, 
 and Middleton, Henry I and Henry II (Davenport, 
 probably) to Shakespeare and Davenport, and The 
 
APPENDICES 179 
 
 History of King Stephen^ Duke Humphrey, and Iphis 
 and lantha to Shakespeare. 
 
 Webster's works have, in one way .and another, been 
 pietty thoroughly scrutinised for parallels. Resem- 
 blances in phrasing and thought between The White 
 Devil, The Duchess of Malfi, The DeviVs Law-Case, and 
 A Monumental Column are very numerous. A Cure 
 for a Cuckold and Appius and Virginia are far less 
 closely joined. In A Cure for a Cuckold there are cer- 
 tain minor echoes of phrase that have some weight. 
 I give a list of the only connections of Appius and Vir- 
 ginia with the other plays that have been discovered 
 previously, or that I have found.^ 
 
 (a) Appius and Virginia, 149: 
 
 I have seen children oft eat sweetmeats thus. 
 As fearful to devour them: 
 
 Duchess of Malfi, 65 : 
 
 I have seen children oft eat sweetmeats thus, 
 As fearful to devour them too soon. 
 
 (&) A. and v., 151: 
 
 One whose mind 
 Appears more like a ceremonious chapel 
 Full of sweet music, than a thronging presence. 
 
 Duchess of Malfi, 79 : 
 
 His breast was filled with all perfection, 
 And yet it seemed a private whispering-room 
 It made so little noise of 't. 
 *The references are all by the pages of Dyce's one-volume edi- 
 tion. 
 
180 JOHN WEBSTER 
 
 Monvmental Column, 11. 78, 79: 
 
 Who had his breast instated with the choice 
 Of virtues, though they made no ambitious noise. 
 
 (c) A. and F., 163: 
 
 Virginia. But she hath a matchless eye. Sir. 
 CoRBULo. True, her eyes are no right matches. 
 
 White Demi, 31 : 
 
 Brachiano. Are not those matchless eyes mine? 
 ViTTORiA. I had rather 
 
 They were not matches.^ 
 
 {d) A. and v., 165: 
 
 I only give you my opinion, 
 I ask no fee for 't. 
 
 Westward Ho! 242: 
 
 Take my counsel: I'll ask no fee for *t. 
 White Devil 7: 
 
 This is my counsel and I'll ask no fee for 't. 
 
 {e) A, and v., 168: 
 
 As aconitum, a strong poison, brings 
 
 A present cure against all serpents' stings. 
 
 White Devil, 26 : 
 
 Physicians, that cure poisons, still do work 
 With counter-poisons. 
 * Quarto reading. Dyce reads "matchless": obviously wrongly. 
 
APPENDICES 181 
 
 (/) A, and v., 171: 
 
 I vow this is a practised dialogue: 
 Comes it not rarely off? 
 
 Duchess of Malfl, 63 : 
 
 I think this speech between you both was studied. 
 It came so roundly off. 
 
 (g) A. andV,, 172: 
 
 For we wot 
 The Office of a Justice is perverted quite 
 When one thief hangs another.^ 
 
 Duchess of Malfi, 90: 
 
 The office of justice is perverted quite 
 When one thief hangs another. 
 
 (h) A. and F., 180: 
 
 Death is terrible 
 Unto a conscience that's oppressed with guilt! 
 
 Duchess of Malfi, 99 : 
 
 How tedious is a guilty conscience! 
 
 (0 A. and F., 173: 
 
 I have sung 
 With an unskilful, yet a willing voice, 
 To bring my girl asleep. 
 
 *So Quarto. Dyce thinks this a mistake for "The office of jus- 
 tice. . . ." as in The Duchess of Malfi quotation. He is probably 
 right. 
 
182 JOHN WEBSTER 
 
 White DevU, 45 : 
 
 I'll tie a garland here about his head; 
 'Twill keep my boy from lightning. 
 
 Besides these, there are various words: "dunghill" 
 (A. and v., 171, 166, White Devil, 25), "mist" (of ig- 
 norance) (A. and F., 167, 170, White Devil, 50 i) are 
 favourite and typical words of Webster. Note also 
 "pursenet" in the sense of "wile" (A. and F., 170, 
 DeviVs Law-Case, 130) and "not-being" {A. and F., 
 180, Duchess of Malfl, 90). 
 
 Of the resemblances, (c) is a common joke, (e) a, 
 common idea (the Ben Jonson, Sejanus, quotation 
 which Dyce gives in a note is much nearer than the 
 passage from the White Devil to the A. and F. quo- 
 tation), and (d) sounds like a catch-phrase. In (h) 
 the two examples occur near the end of their pla3^s, 
 and slightly recall each other in atmosphere. In (i) 
 the same effect of tenderness is got by the word "my." 
 
 It seems to me that (6), a suggestion of Mr. Craw- 
 ford's, holds good only between The Duchess of Malfi 
 and A Monumental Column. 
 
 These six examples are such that they would be 
 important if they were ten or fifteen times as numer- 
 ous ; being so few they are of no account. And I do 
 not think many more could be found. 
 
 The rest, (a), (/) and (g), are another matter. It 
 is to be noted that (a) and (g) are exactly the sort 
 
 * Especially the similarity between "in a mist," A. and V., 167, 
 and "in a mist," White Devil, 50. 
 
APPENDICES 188 
 
 of images and proverbial sayings (note the expression 
 "we wot") that Webster and others collected. If Web- 
 ster wrote Appius and Virginia, we can only say that 
 he must have used the same note-book that he wrote 
 The Duchess of Malfi with. If not, either the author 
 of Appius and Virginia compiled his note-book out of 
 The Duchess of Malfi among other books ; or else they 
 used common sources. (/) is an even more significant 
 parallel. For the circumstances are similar. In each 
 drama two "villains" play into each other's hands in 
 a dialogue which the "hero" discerns, suddenly, or 
 guesses, to have been rehearsed. It is not an obvious 
 thought. That it should be expressed at all is note- 
 worthy ; that it should be expressed with such similarity 
 of phrase and (which is important) metrical setting, 
 is a valuable proof of identity of authorship. 
 
 The words have little weight. The use of "mist" is 
 striking; but "dunghill," though it irresistibly recalls 
 Webster's manner, was not monopolised by him; and 
 "not-being" (the repetition of which Dr. Stoll seems 
 to think remarkable) is not rare enough or typical 
 enough to be of any significance. 
 
 There the proofs of Webster's authorship end. The 
 attribution of a late publisher, which is evidence of a 
 notoriously untrustworthy character, and three or 
 four passages of repetition or resemblance — that is all. 
 The conclusion, for any impartial mind, is that there 
 is very little evidence of the play being Webster's, 
 rather more for his having had a finger in it, but 
 
184 JOHN WEBSTER 
 
 much stronger evidence still that he had practically 
 nothing to do with it. 
 
 Ill 
 
 If that is all there is to be said, we are left with an 
 impression of general confusion, and a strongish feel- 
 ing that anyhow Webster is responsible for very little 
 of the play. 
 
 But the question would be cleared, if anyone dis- 
 covered a more promising candidate. This I believe 
 I have done. I think I can show that Appius and 
 Virginia is largely, or entirely, the work of Thomas 
 Heywood. I shall give the direct proofs first: then 
 the more indirect ones, by showing how his authorship 
 fits in with the various facts that have made such havoc 
 of Webster's claims. 
 
 I have mentioned the queer distinctive vocabulary, 
 especially of Latin words, used in Appius and Vir- 
 ginia. The fact that Heywood uses a very similar 
 vocabulary, especially in all his more classical works, 
 would of itself be of little weight. But an individual 
 examination of all the very unusual words and phrases 
 in this play, together with a hurried scrutiny of Hey- 
 wood's dramas, provides very startling results. I give 
 a list. More minute search, no doubt, might largely 
 increase it. It serves its purpose. I begin with the 
 more striking words. ^ 
 
 * The references to Heywood's plays are to the pages of the six- 
 volume Pearson edition, 1874, 
 
APPENDICES 185 
 
 A. and v., 179: 
 
 Redeem a base life with a noble death. 
 And through your lust-burnt veins confine youi 
 breath. 
 
 "Confine," in this sense of "banish," was very rare. 
 The N.E.D. gives one more or less contemporary ex- 
 ample from Holinshcd, and one, the only one, from 
 Shakespeare. Dyce, in a footnote, gives five passages ; 
 he comments, "it is somewhat remarkable that they are 
 all from Heywood." I can add two. It was a very 
 special word of Heywood's. 
 
 Pleasant Dialogues, ii. p. 115: 
 
 The soul confine. 
 The body's dead, nor canst thou call it thine 
 
 Royal King and Loyal Subject, 82: 
 
 Which as your gift I'll keep, till Heaven and Nature 
 
 Confine it hence. 
 
 It is to be noticed that the context in these two ex- 
 amples is similar. 
 
 Other examples are in The Golden Age, 23, Th£ Rape 
 of Lucrece, 242, A Challenge for Beauty, 10, The 
 Brazen Age, 199, VwaiKetoVj iv, 207. 
 
 A, and V„ 174: 
 
 If the general's heart be so obdure. 
 
 "Obdure" is a very rare word. It does not occur in 
 Shakespeare. In the Elizabethan age it seems to have 
 
186 JOHN WEBSTER 
 
 been used only by one or two religious writers and 
 Heywood. Heywood is always using it. This word 
 alone might almost be accepted as a proof that the 
 passage it occurs in was by him. 
 
 "Obdure" as adjective occurs in Lucrece, 219, 224, 
 Golden Age, 56, 60, Forttme hy Land and Sea, 375, 
 Pleasant Dialogues, 114: as verb, English Traveller, 
 90, VvvaiKeLovj i. 55, Brit. Troy^ vi. 11. "Obdureness" 
 comes in TwaiKtlov, i, 55. 
 
 A. andV.,lQ9.:''Palpedr 
 
 There are only three known instances of this extraor- 
 dinary word; this one, and two from Heywood's ac- 
 knowledged works: Brit. Troy, xv. xlii. and Brazen 
 Age, 206. 
 
 I add a short list of instances that are less per- 
 suasive individually, but have enormous weight collec- 
 tively. 
 
 A. and F., 152: 
 
 Why should my lord droop, or deject his eye.^ 
 
 Rare in this literal sense: not in Shakespeare. Hey- 
 wood. // yoii know not me, 206: 
 
 It becomes not 
 You, being a Princess, to deject your knee. 
 
 Cf. also Lucrece, 173, "dejected," 174, "dejection." 
 
APPENDICES 187 
 
 A. and F., 153, prostrate, in a very uncommon meta- 
 phorical usage: 
 
 Your daughter . . . most humbly 
 Prostrates her filial duty. 
 
 This is paralleled twice in Heywood's The Rape of 
 Lucrece, and once in another play: 
 
 Rape of Lucre ce, 173 : 
 
 This hand . . . 
 
 Lays his victorious sword at Tarquin's feet. 
 
 And prostrates with that sword allegiance. 
 
 Pp. 211, 212: 
 
 The richest entertainment lives with us (i.e. that 
 
 lives with us) 
 According to the hour, and the provision 
 Of a poor wife in the absence of her husband. 
 We prostrate to you. 
 
 Royal King and Loyal Subject, 42 : 
 
 To you ... my liege, 
 A virgin's love I prostrate, 
 
 A,andV.,15S: 
 
 An infinite 
 Of fair Rome's sons. 
 
 "Infinite" is sometimes, though rarely, used by itself, 
 more or less as a number. But used merely as a sub- 
 stantive, as here, it is very unusual. It is found in 
 
188 JOHN WEBSTER 
 
 Hey wood's Rape of Lucrece, 234, Golden Age, 36 ; cf. 
 also Rape of Lucrece, 243 : 
 
 Before thee infinite gaze on thy face. 
 
 A, amd F., 153: 
 
 The iron wall 
 That rings this pomp in from invasive steel. 
 
 A rare word. Once in Shakespeare. The phrase is 
 repeated in Heywood's Golden Age, 40: 
 
 The big Titanoys 
 Plow up thy land with their invasive steel. 
 
 A, and v., 15^: 
 
 Let Janus' temple be devolved (i.e. overturned). 
 
 A very rare word in this sense. The N.E.D, gives only 
 two other examples, one of 1470, one of 1658. Not in 
 Shakespeare. Heywood, Lucrece, 244 : 
 
 For they behind him will devolve the bridge. 
 
 A, and v., 155: 
 
 You mediate excuse for courtesies. 
 {i.e. beg on somebody else's behalf.) 
 
 Rare: not in Shakespeare. In Webster's The White 
 Devil in the sense of "to take a moderate position!" 
 Marlowe and one or two prose-writers have used it in 
 
APPENDICES 189 
 
 the sense of the text. It is found in Heywood, English 
 Traveller, 84: 
 
 Will you. . . . 
 Not mediate my peace? 
 
 A, and F., 161: 
 
 Upon my infallid evidence. 
 
 Very rare: not in Shakespeare. N.E.D. gives only 
 two other examples, of which one is Heywood, Hier- 
 arch., V. 308: 
 
 All these are infallid testimonies 
 
 A,andV„ 174: 
 
 Let him come thrill his partisan 
 Against this breast. 
 
 ^'Thrill, i.e. hurl, — an unusual sense of the word," says 
 Dyce. He adds two quotations, both from Heywood's 
 Iron Age, e.g. p. 316: 
 
 All which their javelins thrild against thy breast. 
 
 Note the correspondence of phrase. This use is not 
 found in Shakespeare. 
 
 A. and v., 174: 
 
 Marshal yourselves, and entertain this novel 
 Within a ring of steel. 
 
190 JOHN WEBSTER 
 
 An uncommon substantive, not found in Shakespeare. 
 Hejwood, English Traveller, 27, Golden Age, 55, Iron 
 Age, Second Part, 373, Brazen Age, 202. 
 
 A. and v., 178: 
 
 One reared on a popular suiFrage 
 Whose station's built on aves and applause. 
 
 For this sense, "shouts of applause," the N.E.D. 
 gives only two examples ; one from Shakespeare {Meas- 
 ure for Measure), the other from Hey wood, Golden 
 Age, 8. 
 
 And all the people with loud suffrages 
 Have shrilled their aves high above the clouds. 
 
 Note the conjunction with "suffrage." The human 
 brain works half mechanical along tiny associative 
 paths; and minute hints of this kind, as a backing to 
 more tangible instances of the uses of very rare words, 
 importantly help this sort of proof. Heywood also 
 uses the word uniquely. Golden Age, 47. 
 
 The people ave'd thee to heaven. 
 
 A.andV,, 179: 
 
 This sight has stiffened all my operant powers. 
 
 Dyee quotes Handet, iii. 2 : 
 
 My operant powers their function leave to do. 
 
 And it is quite probable that the author of Appius and 
 Virgmia is borrowing the phrase from Shakespeare, 
 
APPENDICES 191 
 
 for the word is very uncommon. Heywood, in The 
 Royal King and the Loyal Subject, probably written 
 just about the same time as Hamlet, uses the word, in 
 the same sense (p. 6), only writing "parts" instead of 
 "powers." The sense of this passage is even nearer to 
 the Hamlet line: they are obviously connected — 
 through Heywood, as usual, echoing rather than imi- 
 tating Shakespeare. 
 
 When I forget thee may my operant parts 
 Each one forget their office. 
 
 It seems to me probable that Heywood echoed Shake- 
 speare immediately in The Royal King and the Loyal 
 Subject, and soon after, rather less closely in Appius 
 and Virginia, 
 
 A. and V., 179: Strage, 
 
 A rare Latinism: not in Shakespeare. Heywood uses 
 it in Pleasant Dialogues, iii. and in The Hierarchie?- 
 There are other general verbal resemblances. The 
 kind of word Heywood invents and uses is the same in 
 Appius and Virgi/nia and through the six volumes of 
 his collected "dramatic works." "Eternized," "mon- 
 archizer," "applausive," "opposure" occur in the lat- 
 ter; "imposturous," "enthronized," "donative," in the 
 former. Who could distinguish? In Appius and Vir- 
 
 * The earlier and longer form of this appendix contains about a 
 dozen further instances of verbal similarity, which were omitted 
 in the later version as being rather less striking than those given 
 here, and therefore unnecessary to the argument. Ed. 
 
192 JOHN WEBSTER 
 
 ginia, 178, he invents (possibly adopts) the rare verb 
 "to oratorize." In The English Traveller, 68, he uses 
 the form "to orator." Resemblances of phrase are as 
 numerous, though not so striking. Heywood was too 
 ordinary and too hurried a writer to have much eccen- 
 tricity of phrase. He wrote in the common style of the 
 time, only slightly garnished by a few queer pet words 
 and a certain Latinism of vocabulary. He does not 
 repeat lines and metaphors as many writers do ; only, 
 occasionally, phrases and collocations of words, but 
 these of such a kind as all his contemporaries repeated 
 also. The result is that it is difficult to find parallels 
 of this nature between any of his works. What there 
 are between Appius and Virginia and the rest, there- 
 fore, have more weight than they would have in the 
 case of some other dramatists. 
 
 There is a rather puzzling expression just at the 
 end of Appius and Virginia (p. 180) : 
 
 Appius died like a Roman gentleman. 
 And a man both ways knowing. 
 
 It is, metrically and in a sense, very like a sentence at 
 the end of The English Traveller (p. 94) : 
 
 Dalavill 
 Hath played the villain^ but for Geraldine, 
 He hath been each way noble. 
 
 Cf . also Fortwne hy Land and Sea^ 386 : 
 
 Come ! I am both ways armed against thy steel. 
 
APPENDICES 193 
 
 One of the few points which the author of Appi/iis 
 and Virgmia introduced into the stories of Dionysius 
 and Livy, is the plot to coerce Virginia by refusing 
 the army's pay and forcing Virginius to sell his goods 
 to pay them. In the first act of A Maidenhood Well 
 Lost (espec. iii fF.) Strozza lays much the same plot 
 against "the General" and his daughter, and what en- 
 sues, the army starving and the general paying the 
 soldiers himself, is exactly the same. This shows, at 
 least, that the idea was in Heywood's mind when he 
 was writing A Maidenhood Well Lost. What is more 
 significant is that another idea in the camp-scenes in 
 Appius and Virginia (also original) was in his mind 
 when he was writing The Rape of Lucrece. On page 
 205 the sentry makes the entirely unnecessary remark 
 about his occupation: 
 
 Thus must poor soldiers do; 
 While their commanders are with dainties fed. 
 And sleep on down, the earth must be our bed. 
 
 This is the nnotif of the whole mutiny-scene in Appkis 
 and Virginia (p. 156). See especially the lines: 
 
 I wake in the wet trench, 
 Loaded with more cold iron than a gaol 
 Would give a murderer, while the general 
 Sleeps in a field-bed, and to mock our hunger 
 Feeds us with scent of the most curious fare 
 That makes his tables crack. 
 
 It is obvious that Heywood's mind ran easily into 
 the same trains of thought. Suggest "Camp" to him, 
 
194 JOHN WEBSTER 
 
 and he readily pictures, in his pleasant light water- 
 colours, the starving, cold soldiers suh divo and the 
 general feeding luxuriously and enjoying a bed. In- 
 deed, the parallels of idea with Lucrece are numerous, 
 as one would expect. Heywood felt that a great man 
 of that time was attended by a "secretary." Porsenna, 
 King of the Tuscans, in his tent (Lucrece, M5) wants 
 lights. He calls "Our Secretary!" The secretary ap- 
 pears with "My lord?" In Appius and Virginia (159, 
 160) when Appius is bearded by Icilius, he calls out 
 for help, "Our Secretary !" and summons him again at 
 the end of the interview, "Our Secretary! . . . We 
 have use for him." Marcus appears: 
 
 My honourable lord.^ . . . 
 
 There are other such small points — the bearing of 
 the dead, bleeding bodies of Lucrece, and of Virginia, 
 before the people, and their sympathy and rage; the 
 vagueness of locality in each play ; and so on. 
 
 But there is a more remarkable resemblance. It is 
 part of a general link with Heywood's works — the 
 clown. Dr. Stoll has three pages (197-200) pointing 
 out and illustrating the kinship of Corbulo in Appius 
 and Virgi/nia with Hey wood's clowns, and especially the 
 clown of The Rape of Lucrece} The Heywood clown, 
 an early type, was a simple, good-hearted creature, who 
 had little to do with the play, and poured out puns and 
 somewhat Euphuistic jokes to amuse the crowd. There 
 
 ^ See also Eckhardt, Die lustige Person irri dlteren englischen 
 Drama, p. 433, etc. 
 
APPENDICES 195 
 
 was a painstaking, verbal tumbling they all indulged in. 
 You can pick at random. "If they suddenly do not 
 strike up," says Slime of the lingering musicians,^ "I 
 shall presently strike them ^ down." It is the voice of 
 Corbulo. The clown in The Golden Age is precisely the 
 same. So is the one in Lucrece, and as the plays are 
 more alike, the similarity of his position is the more 
 easily seen. It is, in the first place, a very remarkable 
 coincidence that he should be there at all. Appius and 
 Virginia and The Rape of Lucrece are the only Roman 
 plaj^s of the adult Elizabethan drama to introduce such 
 a character. It was exactly like Heywood to modify 
 the tradition and genus in this way. It would not 
 have been at all like Webster. Dr. Stoll emphasises 
 and details this similarity so admirably, and as he 
 has no idea that Appius and Virginia is not by Web- 
 ster, his testimony is so valuable in its impartiality, 
 that I cannot do better than quote his description. 
 
 In both cases the clown is servant to the heroine, and he 
 appears in like situations. He is sent by his mistress on 
 errands, is taken to task by her for ogling at her maid (and 
 that in the latter's presence), and is left to chatter with 
 other servants alone. He jokes about his mistress's mis- 
 fortune, about the sinners in the suburbs, and, being a Ro- 
 man, out of the Latin grammar. And the comic side of both 
 is the same. It lies all in the speeclies — the clown plays no 
 pranks and suffers no mishaps — and it has an episodic, 
 random, and anachronistic character. It is all jest and rep- 
 artee, puns, quibbles, and catches, and those neither clever 
 nor new; and the drift of it all, whenever it gets beyond 
 
 ^A Woman killed with Kindness, 97. 
 
 *01d Text "thee!" 
 
196 JOHN WEBSTER 
 
 words, is satire on London life and manners. It is good- 
 humoured, moreover, naive and dirty. 
 
 The episode between the maid, or nurse, and the 
 clown, an entirely irrelevant excrescence, is especially 
 noteworthy. There is even a certain similarity in 
 phrasing and thought, of a kind that suggests the same 
 mind working at different times, rather than imitation. 
 Virginia and Lucrece both address the clown as "Sir," 
 impatiently. Virginia begins: 
 
 You are grown wondrous amorous of late; 
 Why do you look back so often? 
 Lucrece. Sirrah, I ha' seen you oft familiar 
 
 With this my maid and waiting-gentlewoman, 
 As casting amorous glances, wanton looks. 
 And privy becks, savouring incontinence. 
 
 Dr. Stoll, supposing Appius and Virginia Webster's, 
 can of course only suggest that Webster, imitating 
 Shakespeare in the general conception of his play, 
 turned suddenly, picked out one favourite character 
 of Heywood's, and, with Heywood's authority for the 
 anachronism, introduced an extraordinarily good imi- 
 tation of it into his own work. He is like a ventrilo- 
 quist who has at least two lay-figures, each talking 
 with a different voice from the other's, and from their 
 master's. "Eclecticism" is a mild word for such a 
 method. 
 
 IV 
 
 Anyone who believes in Webster's authorship of 
 the play, has now got to explain away not only the 
 
APPENDICES 197 
 
 date difficulty, not only the general aesthetic absurdity, 
 not only the borrowing of a pet character of Hey- 
 wood's, but also the sudden entire adoption of Hey- 
 wood's individual, distinguishing vocabulary. Twenty 
 years' friendship, you are to suppose, never affected 
 Webster's vocabulary in this direction in the slightest 
 degree. Then, in a transport of "senile" affection, he 
 hurled aside his own personality, and became mere 
 Tom. 
 
 In the next place, consider how the theory of Hey- 
 wood's authorship suits the facts of the play. If 
 Hey wood wrote Appius and Virginia, there is no diffi- 
 culty about words or handling. He wrote the play 
 most like it of all the plays in the world. There is 
 no difficulty about style. It is exactly like Heywood 
 when he is writing solemnly, as in parts of Lucrece, 
 parts of the various "Ages," and the beginning and 
 end of The Royal King and the Loyal Subject. Only 
 it is rather more mature, it has a little more freedom 
 and rhetoric, than the early style of Lucrece and some 
 of the "Ages." This suits the other indications of 
 date. For, again, there is no difficulty about the date. 
 The difference between Lucrece and Appius and Vir- 
 ginia is mostly due to the fact that Coriolanus (c. 
 1608) must have intervened. Any date after 1608 
 would do ; immediately after is the most likely, because 
 the resemblances of style and vocabulary are, on the 
 whole, to the rather earlier works. 
 
 I imagine- that the main part of Appius and Vir- 
 ginia, as we have it, was written then. It may, and 
 
198 JOHN WEBSTER 
 
 indeed must, have been cut about and altered, by Hey- 
 wood or others, before it found a last home with "Bees- 
 ton's boys" in 1639, or a final resting-place with Mose- 
 ley in 1654. 
 
 The metrical characteristics noticed in Appius and 
 Virginia are Heywood's. Heywood's blank verse, says 
 Dr. Schipper,^ is "sehr gewandt und harmonisch ge- 
 baut." This applies perfectly to our play. He also 
 calls attention, of course, to the number of rhyming 
 couplets, ending off even short speeches. It is this 
 characteristic in Appius and Virginia that slightly puz- 
 zles Dr. Stoll and suddenly upsets his metrical tables 
 (p. 190). The only detailed examination of Hey- 
 wood's prosody that I know is in Dr. Franz Albert's 
 "tJber Thomas Heywood's Life and Death of Hector 
 of Trot/.'* ^ It is concerned mainly with certain sides 
 of Heywood's work, mostly undramatic, and it is not 
 very perspicacious, having most of the faults of Ger- 
 mans trying to understand English metre. But it 
 enumerates some of the more tangible characteristics, 
 and lays great stress on that trick of conscious and 
 rather conventionalised elision, especially between "to" 
 and a verb with an initial vowel, that I had already in- 
 dependently noticed in Appius and Virginia^ and have 
 remarked on earlier in this appendix. 
 
 The various characteristics of the play that are 
 no bar to Webster's authorship fit in equally well or 
 better with Heywood's. This is the case with the 
 
 ^Englische Metrik, 1881, vol. ii. p. 335. 
 ^ Especially pp. 22, 112. 
 
APPENDICES 199 
 
 numerous slight imitations of phrases of Shakespeare, 
 which are rather more a mark of Heywood than of 
 Webster.-^ 
 
 The sources of Appius and Virginia ^ are, ultimately, 
 Livy and Dionysius of Halicarnassus. Dr. Lauschke 
 believes he used both of these, and also Painter, who 
 paraphrased Livy, and Giovanni Fiorentino, the Ital- 
 ian translator of Dionysius. As Dr. Stoll points out, 
 there is no evidence for Giovanni Fiorentino, and very 
 little for Livy in the original, as against Painter.^ 
 They do not seem, however, to have considered the pos- 
 sibility of Philemon Holland's well-known translation 
 of Livy (1600). In the passage where the question of 
 Virginia's custody till the trial is being discussed, Hol- 
 land introduces the technical legal word "forthcom- 
 ing." Appius and Virginia makes good use of the word 
 in the corresponding passage (p. 167). Painter does 
 not use it, and the Latin does not necessarily suggest 
 it. The author of Appius and Virginia may have 
 thought of it for himself, in reading the original. But 
 it decidedly points to Holland being used; and there- 
 fore does away with the necessity of either Painter or 
 
 ^See Cambridge History of English Literature, vol. vi. p. 106. 
 
 ' See Lauschke, "John Webster's Tragodie Appius and Virginia/' 
 and Stoll, pp. 160-162. 
 
 'There are two points: (<x) Livy has "sordidatus" ; A. and V. 
 "disguised in dust and sweat"; Painter nothing. This is very 
 little, and becomes nothing when you realise — Dr. Stoll does not 
 point it out, though Lauschke does — that "sordidatus" and "dis- 
 guised . . ." come in entirely different parts of the story. (6) 
 Minutius as the name of the general at Algidum occurs in Livy, 
 not in Painter or Dionysius. This has a little weight. 
 
200 JOHN WEBSTER 
 
 Livy. It is certain that Dionjsius was used,^ in the 
 original or a Latin translation (there was probably 
 no English translation at this time). The sources, 
 then, favour Heywood if anything. Of Webster's 
 classical knowledge we can only say that he knew other 
 people's Latin quotations. Thomas Heywood, Fellow 
 of Peterhouse, translator of Sallust, Ovid, and Lucian, 
 author of the learned Hier archie. Apology for Actors, 
 TvvaLKehv, etc., was a lover of learning and a reader 
 of Latin and Greek all his life. 
 
 It remains to see what explanation, on the assump- 
 tion that Heywood is mainly or entirely the author of 
 Appius and Virginia, can be given of the exiguous 
 pieces of evidence that point towards Webster. There 
 is first Moseley's attribution. I have said how little 
 weight the attribution of a late publisher carries. In 
 this case it is impossible to do much more than theorise 
 about what can have happened. If Heywood's name 
 was on the play when Moseley got it, it is unlikely he 
 would have changed it for Webster's, not only because 
 he seems to have been fairly honest, but also because 
 there was not sufficient inducement. Of the two, how- 
 ever, Webster was the more famous and attractive after 
 the Civil War. Winstanley (1686) (who— it is an odd 
 accident — mentions all Webster's plays except Appius 
 * V. Stoll, p. 162, for conclusive proofs. 
 
APPENDICES 201 
 
 avd Virginia) makes little of either of them. Phillips 
 (1674) says Webster was the author of "several not 
 wholly to be rejected plays"; on the identity of which, 
 however, he was terribly shaky. Heywood he dismisses 
 even more cursorily as the writer of "many but vulgar 
 comedies." Langbaine, who always takes a rather high 
 tone, describes Webster as "an author that lived in 
 the reign of King James the First, and was in those 
 days accounted an excellent poet." But he goes on to 
 confess that The Duchess of Malfi, The White Devil, 
 and Appius and Virginia, "have, even in our own age, 
 gained applause." It was true. The White Devil was 
 being acted at the Theatre Royal in 1671, and a quarto 
 of it was printed in the following year. The Duchess 
 of Malfl was acted in 1664 at Lincoln's Inn Fields, and 
 in 1667 at the Duke's Theatre. It was reprinted in 
 the same year. Downes {Roscius Anglicanus) de- 
 scribes it as "one of the best stock-tragedies." Appius 
 and Virgvnia, as Webster's, with Betterton's alterations, 
 was acted at the Duke's Theatre in 1670. Mrs. Better- 
 ton was Virginia. Genest quotes from Downes that 
 it ran for eight days, and was very frequently acted 
 afterwards. All this shows that Webster's name was 
 fairly well known in this period. There is no trace of 
 any known play of Heywood's being revived. 
 
 It is easy enough to imagine a play of his coming 
 without a name, or with a wrong name, into the hands 
 of a pubhsher of 1654. There were two hundred and 
 twenty plays "in which I have had either an entire 
 
202 JOHN WEBSTER 
 
 hand or at least a main finger." ^ On any that came 
 to the press in his hfetime, he seems to have kept an 
 eye. For the others, when they had passed out of his 
 control, he seems not to have cared. "Many of them, 
 by shifting and change of companies have been negli- 
 gently lost; others of them are still retained in the 
 hands of some actors who think it against their pecu- 
 liar profit to have them come in print." ^ Appius and 
 Virginia may have belonged to either, more probably 
 to the latter class. And it is very easy to trace a pos- 
 sible and probable history of this play.^ We first hear 
 of it in 1639, in the possession of Christopher Beeston's 
 company of boys, who occupied the Cockpit Theatre 
 from 1637 onwards. Now Christopher Beeston and 
 Thomas Heyi70od were members of Queen Anne's com- 
 pany from its foundation in 1603. In 1617 the Cock- 
 pit opened, and Queen Anne's company went there till 
 1619. From 1619 to 1625 the lady Elizabeth's com- 
 pany held the Cockpit, and probably, though not cer- 
 tainly, Heywood and Beeston were of them. From 
 1625 to 1637 they were followed by Queen Henrietta's 
 company, managed by Beeston. And then came Bee- 
 ston's company of boys, who possessed the play in 
 1639. Among all the various strands of continuity in 
 the Elizabethan theatres and companies, this is a very 
 definite one, forming about Heywood and Beeston, in 
 connection first with Queen Anne's company, and then, 
 
 ^Tke English Traveller: To the Reader. ^ Ibid. 
 
 ' See Murray, English Dramatic Companies, vol. i. pp. 265-370, 
 and elsewhere. 
 
APPENDICES 203 
 
 locally, with the Cockpit. And with Heywood, Bee- 
 ston, and, I believe, Appius and Virginia, on this long 
 journey, goes significantly The Rape of Lucrece, 
 
 It is also to be noticed that it was Queen Anne's 
 company that acted two of Webster's three original 
 plays. The White Devil (1611) and The Devil's Law- 
 Case (1620). He seems to have gone off to the King's 
 Men between these, with The Duchess of Malfi (1612- 
 1613). But we may suppose that he had most to do 
 with Queen Anne's company. 
 
 There remain the similarities and repetitions of 
 phrase in Appius and Virginia and Webster's plays. 
 As I have said, only three of these are of any im- 
 portance, two exact verbal repetitions and one strik- 
 ing similarity of phrase and idea; all connecting with 
 The Duchess of Malfi} If Heywood wrote the main 
 part or all of Appius and Virginia, there are six pos- 
 sible explanations of these passages. They are an 
 accident; or Heywood imitated Webster; or Webster 
 imitated Heywood; or the play was touched up by 
 some Queen's company actor or author who knew The 
 Duchess of Malfi; or Webster himself touched it up; 
 or Webster and Heywood wrote Appius and Virginia 
 together, Heywood taking the chief part. 
 
 The first is improbable, though far less improbable 
 than it seems. For both {a) and {g) are sententious 
 sayings such as the Elizabethans delighted to note 
 down and repeat. Webster is full of these. And the 
 identical repetition of one of them by him and Marston 
 *(«)» (/)» and {g) in my list (pp. 179-182). 
 
204 JOHN WEBSTER 
 
 supported great theories of his imitation of Marston 
 till Mr. Crawford discovered it in Montaigne/ the 
 common source to which thej had independently gone. 
 Still, the coincidence of the two apophthegms is rather 
 much to account for in this way. It is possible, but 
 that is all. And there is the further difficulty against 
 it that Heywood was not wont to write in this note- 
 book manner. He worked too quickly. 
 
 This also counts against what might otherwise seem 
 an easier theory, that (/) is either an accident or the 
 imitation of reminiscence, but that these two (a) and 
 (g) are the result of Heywood directly copying Web- 
 ster — noting down and using two of his phrases. The 
 possibility of tliis is also lessened by the probability 
 on other grounds that Appius and Virginia is earlier 
 than The Duchess of Malfl. Webster may have imi- 
 tated Heywood. He was a great friend of his at this 
 time.^ And if Appius and Virginia was, as is prob- 
 able, written early, it must have appeared in the same 
 theatre and about the same time as The White Devil,^ 
 Also it was Webster's habit to take down from other 
 authors and afterwards use sentences and similes of an 
 apophthegmatic or striking nature. We know that 
 he treated Donne, Montaigne, Jonson, Sidney, and 
 perhaps Marston and Dekker in this way. Why not 
 
 * Crawford, Collectanea, Series ii. p. 35. 
 
 ' He wrote some lines "To his beloved friend Master Thomas 
 Heywood," prefixed to Heywood's Apology for Actors, 1612. 
 
 ' It is an important indication of the date of Appius and Vif' 
 ginia that The White Devil (1611) does not borrow from it, and 
 The Duchess of Malfl (1612-13) does. 
 
APPENDICES 205 
 
 Heywood, his friend and collaborator ? It is true Hey- 
 wood does not lend himself often so easily to such use. 
 That, and the fact that he has not been thoroughly 
 searched for such a purpose, may explain why there 
 are few other known parallels. This theory is the more 
 probable because the lines of (a) and (g), and their 
 ideas, seem more natural and in place in Appius and 
 Virginia than they do in The Duchess of Malfl. And 
 it is easier to imagine Webster finding {Appius and 
 Virginia, 149), 
 
 I have seen children oft eat sweetmeats thus, 
 As fearful to devour them. 
 
 and adding (Duchess of Malfl, 65) the words "too 
 soon" than Heywood doing the opposite. 
 
 There remain the various possibilities of two hands 
 having been at work, or the same hand at two periods. 
 These are favoured by the a priori probability of a 
 play that had at least thirty years of acting life being 
 altered in the period, and also by certain indications 
 that all is not right with the play as it stands. These 
 I shall shortly set out. 
 
 In the beginning of Act I. there is a queer and soli- 
 tary passage of prose which looks like an abbreviation 
 for acting purposes. Dyce suspects it ; and it is to be 
 noted that the speech following the prose contains one 
 of the two "repetitions" from The Duchess of Malfi. 
 
 In II. 3 (p. 160) there are difficulties which seem 
 to have passed unnoticed. Icilius comes to plead with 
 
206 JOHN WEBSTER 
 
 Appius for the camp, and so for Virginius. Appius 
 counters with a proposal that Icilius should give up 
 Virginia, and marry into his own family. Icilius flies 
 out with the charge that Appius has been lustfully 
 tempting Virginia with presents and letters. Appius 
 is prevented by force and threats from either calling 
 for help or replying. At length the storm subsides. 
 Appius replies, pretending he knows nothing of it, 
 playing indulgent eld. Icilius crumbles completely. 
 
 I. I crave your pardon. 
 
 A. Granted ere craved, my good Icilius. 
 
 I. Morrow. 
 
 A. It is no more indeed. Morrow, Icilius, 
 
 If any of our servants wait without, 
 
 Command them in. 
 
 I do not think any good sense can be made out of that 
 "It is no more indeed." It looks, at first sight, like a 
 pun on "morrow." But that does not help. Indeed the 
 whole collapse of Icilius is oddly curt and sudden. It 
 seems to me probable that a cut has been made here, 
 or some other operation of hasty revision. 
 
 And in the next scene, III. 1 (pp. 161-2) Icilius 
 reports the interview to his friends and Virginia. He 
 went, he says, to Appius, took him by the throat, 
 forced him to hear, taxed him with his lust and his 
 behaviour, "with such known circumstance" that Ap- 
 pius could try to excuse it, but could not deny it. 
 They parted "friends in outward show" ; Appius swore 
 "quite to abjure her love"; but yet had continued his 
 messages. 
 
APPENDICES 207 
 
 Now this is quite a different story from the truth. 
 In a play of this kind, simple in characterisation and 
 full of childishness in construction and episode, we 
 cannot suppose the author was attempting the subtle 
 irony Ibsen practised in The Wild Duck, where you 
 see the truth in one scene and Hialmar Ekdal's family 
 version of it in the next. Nor would such a sudden 
 spasm of Euripidean double-dealing help either the 
 character of Icilius or the play. Besides, there are 
 other indications of confusion. For when (III. 2, p. 
 164) Virginia is suddenly arrested, she cries out: 
 
 O my Icilius, your incredulity 
 Hath quite undone me ! 
 
 which seems to refer to the first, true version of the 
 story, and to mean that Icilius' not believing her but 
 accepting Appius' defence had ruined her. These 
 seem to me to be plain signs that the scenes as they 
 stand have been written, to some extent at least, by two 
 people, or by the same person at different times. 
 
 Another discrepancy affecting the same point, the 
 interview and the report of it, is mentioned by Dyce 
 in his note on 11. 3 (p. 158). The scene would seem 
 to be an outer apartment in the house of Appius. 
 But presently, when Appius is left alone with Icilius, 
 a change of scene is supposed: for he says to Claudius 
 (p. 160): 
 
 To send a ruffian hither. 
 Even to my closet! 
 
208 JOHN WEBSTER 
 
 And jet, in the first scene of the next act, Icilius 
 speaks of the interview as having taken place in the 
 lobby! 
 
 The only other suspicion of corruption in this play 
 which I know of may as well be mentioned here. Mr. 
 Pierce ^ believes that III. 4, the conversation between 
 Corbulo and the serving-men, was interpolated to please 
 the groundlings. His reasons are: (1) it is wholly in 
 prose; (2) the doggerel rhyme; (3) it does not ad- 
 vance the action; (4) the average number of three- 
 syllable Latin words (his particular test) is very low. 
 I do not feel convinced. The scene is extremely Hey- 
 woodian. The Latin-word test is not so important as 
 Mr. Pierce appears to think, especially when applied 
 to a short, rather comic, prose-scene. And it affects 
 Heywood far less than Webster. No doubt this scene 
 was put in "to please the groundlings." But it was 
 put in by the author. 
 
 The conclusion, then, that the play as we have it 
 has been revised and altered, helps any theory that 
 Webster and Heywood each had a finger in it. It 
 might, of course, have been changed by any member 
 of the Queen's Servants' Company. But he would 
 not be likely to have incorporated passages from The 
 Duchess of Malfi, a play belonging to the King's Men. 
 If it was Heywood himself that touched it up, in 1613 
 or so, he might quite well have done this, being a friend 
 of Webster's. But it is most easy to suppose Webster 
 the reviser. Either this, or his collaboration, is ren- 
 * The Collaboration of Webster and Dekker. 
 
APPENDICES 209 
 
 dered rather probable by the presence through the 
 play of ten or a dozen passages, averaging perhaps two 
 lines, that seem to taste slightly of his style. Perhaps 
 it is true that any play, examined closely, would yield 
 the same. And certainly Heywood covld have written 
 them. But, at moments, there does seem to be the 
 flavour, almost imperceptibly present. If reviser or 
 collaborator, Webster obviously had recourse to the 
 same note-books as he used for The Duchess of Malft, 
 which suggests that he would be working on it about 
 161S or soon after. And in either case, we should 
 have a very good explanation of his name being con- 
 nected with the play. If he revised, we must sup- 
 pose that he shortened and made more dramatic the 
 very beginning of the play, and heightened, or even 
 rewrote, the trial scene (IV. 1). It is important to 
 notice that in this rather long scene (1) there are no 
 very characteristic words of Hey wood's, (2) there are 
 more of the phrases, words and lines that are faintly 
 reminiscent of Webster than anywhere else in the play,^ 
 (3) two 2 of the three strong indications of a connec- 
 tion with Webster occur. 
 
 Give Webster the revision of these two scenes, and 
 you have satisfied his utmost claims. To yield him 
 more is mere charity. If he collaborated, it is impos- 
 sible to divide the play up between the two. In certain 
 
 1 "Dunghill," "mist," "pursenet," "to bring my girl asleep," "and 
 this short dance of life is full of changes," etc., etc. 
 U. e. if) and (^r). 
 
210 JOHN WEBSTER 
 
 scenes (e.g. IV. 2 and V. 3) Hey wood's vocabulary 
 comes out more clearly than in the rest. But one can 
 only say that Webster's part is very small compared 
 with Heywood's, as unimportant as it is in Northward 
 Ho and Westward Ho. 
 
 In sum: general, critical, and aesthetic impressions, 
 more particular examination of various aspects, and 
 the difficulty of fitting it in chronologically, make it 
 impossible to believe that Appius and Virginia is by 
 Webster, while the evidence in favour of his authorship 
 is very slight. All these considerations, and also re- 
 markable features of vocabulary and characterisation, 
 make it highly probable that it is by Heywood. The 
 slight similarities between The Duchess of Malfl and 
 Appius and Virginia may be due to Webster borrowing 
 in The Duchess of Malfi from Heywood, or revising 
 Appius and Virginia, or having, not for the first time, 
 collaborated with Heywood, but very subordinately. 
 In any case, Appius and Virgima must be counted 
 among Heywood's plays ; not the best of them, but 
 among the better ones ; a typical example of him in 
 his finer moments, written rather more carefully than 
 is usual with that happy man. 
 
Appendix B. — Miscellaneous 
 
 NON-EXTANT PLAYS 
 
 There are no difficulties about the dates of most of 
 the non-extant plays. Ccesar's Fall, Two Shapes^ and 
 Christmas Comes but Once a Year are dated 1602 by 
 the entries in Henslowe. Dr. Greg from the list of 
 collaborators and the nearness in date of the payments 
 thinks Ccesar^s Fall and Two Shapes must be the same 
 play ; it may be so, but it is not convincing. Henslowe 
 may very well have been employing the same people 
 in the same month to write two plays. There is a 
 doubt about the name of Two Shapes. That is Dr. 
 Greg's reading. Collier read Two Harpes; which some 
 construe Two Harpies. 
 
 A Late Murther of the Son upon the Mother by Ford 
 and Webster is entered in Herbert's Office Book for 
 September, 1624. Pamphlets of July, 1624, about 
 such a murder case are on record. The play must 
 have been written in that year. 
 
 The Guise, which Webster mentions in his Dedica- 
 tion to The DetriVs Law-Case, is of quite unknown date. 
 An entry in Henslowe for 1601 giving Webster a play 
 of that name turns out to be a forgery of Collier's. 
 The original entry probably referred to Marlowe's 
 
 211 
 
212 JOHN WEBSTER 
 
 Massacre at Paris, Dr. Stoll, scenting Marlowe In 
 Webster's latest plays, has spun a theory of Webster 
 reading up Marlowe, especially the Massacre at Paris, 
 in his old age. He deduces that we can date Guise 
 about 1620. The whole theory rests on a quite wild 
 assumption that an Elizabethan dramatist, wishing to 
 write a play on a certain subject, began by reading 
 up all previous plays on that subject, like a professor 
 of English Literature. If Webster's own list of plays 
 is in chronological order. Guise is later than 1614. 
 We can say no more. 
 
 The Thracian Wonder, like A Cure for a Cuckold, 
 was first published in 1661 by Francis Kirkman as by 
 Webster and Rowley. No one believes it to be by 
 either. The reasons of this disbelief are entirely aes- 
 thetic. It is dangerous, as I have said elsewhere, to 
 take it for granted that a bad play cannot be by a 
 good author. It is conceivable that Webster and Row- 
 ley might have written or helped to write a play like 
 this at the beginning of their careers. Each has been 
 concerned in equally bad work. But if they did write 
 it, it does not increase our knowledge of them; and if 
 they did not write it, it does not matter who did. 
 So the affair is not very important. A rather unsuc- 
 cessful attempt has been made to explain Kirkman's 
 attribution. Another Webster in 1617 wrote a story, 
 which had no connection with this play, but which 
 
APPENDICES 213 
 
 Kirkman may have thought had. It is not necessary. 
 Kirkman was one of the wildest of the Restoration 
 pubHshers. The fact that he was piibhshing one play 
 as by Webster and Rowley might quite likely lead him 
 to put their names on the title-page of its twin. Any- 
 how he has no authority. We do not know who did or 
 who did not write The Thracian Wonder, 
 
 Monuments of Honour is a quite ordinary city 
 triumph, there is nothing remarkable or important 
 about it. It was published in 1624 as by "John Web- 
 ster, merchant taylor." "John Webster" was a com- 
 mon enough name, and there is no proof that this one 
 is our author. The Latin tag on the title-page, which 
 also ends the preface to The WhiteDevil, was in common 
 use. There is only the probability that no other John 
 Webster would have been distinguished enough in liter- 
 ature to have been chosen to write this. The guilds 
 generally liked to get hold of some fairly accomplished 
 literary man for such a purpose. Neither the verse nor 
 the invention of this pageant affirms the authorship of 
 Webster. But there is also nothing to contradict it. 
 
Appendix C. — Sir Thomas Wyatt 
 
 "the famous history of sir THOMAS WYATT" 
 
 Date. 
 
 The Famous History of Sir Thomas Wyatt. With 
 the Coronation of Queen Mary and the Coming In of 
 King Philip. Written by Thomas Dickers and John 
 Webster, was printed in 1607.^ In October, 1602, 
 Chettle, Dekker, Heywood, Smith, and Webster were 
 paid, in all, £8 for Part I. of Lady Jane or The Over- 
 throw of Rebels; and Dekker was paid, in earnest, 5s. 
 for Part II. (Smith and Chettle may have received 
 small amounts for this, also.) All this was on behalf 
 of Worcester's Men, who passed under the patronage 
 of Queen Anne in 1603. As the 1607 Quarto of Sir 
 Thomas Wyatt says it was played by the Queen's 
 Majesty's Servants, and as the authors are the same, 
 there is no reason to doubt that Dyce was right in 
 supposing that Sir Thomas Wyatt consists of frag- 
 ments of both parts of Lady Jane. Dr. StoU thinks 
 perhaps we have only Part I., as The Coronation of 
 Queen Mary and The Coming In of King Philip are 
 only promised and not given. Dr. Greg suggests that 
 the cut version of Part I. ends and Part II. begins, 
 
 * V. Greg. Henslowe's Diary, Pt. ii. pp. 232, 3. There was an- 
 other edition in 1612. 
 
 214 
 
APPENDICES 215 
 
 with Mary's audience (p. 193, column 2; Scene 10). 
 Professor Schelling makes the credible suggestion that 
 the censor had cut out a great deal; especially, no 
 doubt, the Coming In of King Philip. As it stands, 
 the play is extraordinarily short. In any case, the 
 date is 1602. It must have been played at "The Rose" ; 
 and, as there are two editions, it was probably revived. 
 
 Sources. 
 
 The source of Sir Thomas Wyatt — that is, of the 
 two parts of Lady Jane — is Holinshed ; and, as far as 
 we know, nothing else.^ 
 
 CoUahoration. 
 
 Opinions have differed as to the respective amounts 
 contributed by Dekker and Webster. Dr. Stoll, argu- 
 ing from metre, sentiment, style, phrases, and the gen- 
 eral nature of the play, can find Dekker everywhere, 
 Webster nowhere. Dr. Greg gives Webster rather more 
 than half, mostly the first half. Mr. Pierce ^ says that 
 Webster wrote "most of Scenes 2, 5, 6, 10, 14, and 16, 
 although some of these scenes were certainly retouched 
 by Dekker, and all of them may have been." I shall 
 discuss Mr. Pierce's method of assigning scenes more 
 closely in the Appendix on Westward Ho and North- 
 ward Ho. In the case of Sir Thomas Wyatt none of 
 his metrical tests seems to me to have any validity. 
 
 * V. Stoll, p. 45. 
 
 ' The Collaboration of Webster and Dekker. I use his division 
 into scenes, which is the same as Fleay's. 
 
216 JOHN WEBSTER 
 
 They depend, like Dr. StoU's, on the assumption that 
 Webster's metrical characteristics were the same in 
 1602 as in 1610 or 1620 — an assumption Mr. Pierce 
 himself confesses to be absurd. It must be recognised 
 that we have only three plays on which we can base 
 our generalisations about Webster's metre, two slowly- 
 written Italian tragedies of about 1610 or 1612 and 
 a tragi-comedy of 1620. In Sir Thomas Wyatt Web- 
 ster was writing a different kind of play, together with 
 a lot of other people, probably in a great hurry; and 
 it is likely he was immature. To take the statistics 
 for rhyme in The Duchess of Malfi and the other plays 
 and use them, as proving that Webster uses rhyme less 
 than Dekker, to apportion the scenes in Sir Thomas 
 Wyatt, is a glaring example of that statistical blind- 
 ness and inert stupidity that has continually spoilt the 
 use of the very valuable metrical tables that have been 
 prepared for EHzabethan Drama. The evidence that 
 metre gives in Sir Thomas Wyatt can only be of the 
 vaguest description. 
 
 So, too, with characters. The reason why there are 
 certain kinds of character and incident in any of these 
 three partnership plays, is not that Dekker wrote 
 them. It is that they are that kind of play. If Web- 
 ster wrote a citizen's-wife-gallant play, he must have 
 introduced citizens' wives and gallants, even if he did 
 not do so in an Italian tragedy. On page 2 of his 
 book Mr. Pierce claims that his study is useful as 
 throwing light on Webster's range as an author. "If 
 Webster wrote ... the parts of Captain Jenkins and 
 
APPENDICES 217 
 
 Hans Van Belch in Northward Ho, then he showed an 
 element of pleasant humour and manysidedness which 
 is not indicated anywhere else." In Chapter VII., 
 dealing with "The Character and Atmosphere-Test," 
 he quotes with approval, as proof of what is and what 
 is not Dekker's, Dr. Stoll on these characters. "Mani- 
 festly Dekker's too are the Dutch Drawer and Mer- 
 chant, and the Welsh Captain. A Dutch Hans had 
 already appeared in the Shoemaker . . . and Captain 
 Jenkins ... is the counterpart of Sir Vaughan ap 
 Rees in Satiro-Mastix'' That is to say, these charac- 
 ters of common types are Dekker's, because Dekker 
 uses similar ones elsewhere, and not Webster's because 
 Webster doesn't. You start out to see if Webster, 
 having written only in a certain style elsewhere, wrote 
 in another style here. You conclude that he has not 
 written in this other style here, because he has written 
 only in a certain style elsewhere! 
 
 Considerations of style (in the narrower sense of 
 literary individuality) and vocabulary are more con- 
 vincing. The only one of Mr. Pierce's tests that has 
 any value in the case of Sir Thomas Wyatt — except, 
 of course, the parallel-passages, taken with caution — 
 is his three-syllable-Latin-word one.-^ A large propor- 
 tion of Latin words, and any other characteristic we 
 recognise clearly as one of the later Webster's, do tend 
 to prove his presence in a scene — though their absence 
 does not disprove it. These slight indications of style, 
 if they had arisen and become unconscious so early, 
 
 * See the Appendix on Westwa/rd Ho and Northward Bo, 
 
218 JOHN WEBSTER 
 
 are the things that would be apparent in plays of dif- 
 ferent species by the same author. But the eight or ten 
 years, and the probable presence of so many authors 
 in this play, must make us sceptical. The latter point,' 
 indeed, would falsify most of Mr. Pierce's work if it 
 were sound on other grounds. He remembers, on his 
 last page, that Heywood, Chettle, and Smith also have 
 to be accounted for. He dismisses them too magnifi- 
 cently. "It would be useless to discuss such questions 
 as these at present, since no practical results could 
 follow. We have offered such evidence as we possess 
 on the shares of Dekker and Webster; and here we 
 stop." But though you may not have "discussed" the 
 question of the relative shares of C, D., and E., in a 
 play, you have definitely answered it, if you say A. 
 wrote six scenes and B. the rest. The Latin-word test 
 is no good unless we have Heywood's, Chettle's, and 
 Smith's figures, as well as Dekker's and Webster's. It 
 does not prove that Dekker wrote certain scenes and 
 Webster did not, to say that Dekker employs a "sweet 
 personal tone," or a market-girl with her eggs, else- 
 where, and Webster does not. You have to be able to 
 say that Heywood and Chettle and Smith also are 
 strangers to these things. 
 
 Miss Mary Leland Hunt, in her careful and useful 
 monograph on Dekker,^ also discusses the question of 
 the partition of this play. Her most original sugges- 
 tion is that the main plan of the play is due to Chettle. 
 She advances various indications of this ; that he was 
 * Thomas Dekker: A Study, by Mary Leland Hunt. 
 
APPENDICES 219 
 
 older than Dekker (and Webster, no doubt); that 
 Henslowe mentions his name first ; that he was specially 
 at home in the chronicle history; and that he is more 
 old-fashioned — and so more likely to have planned the 
 old-fashioned structure of Sir Thomas Wyatt — than 
 Dekker. Against Dekker and Webster this certainly 
 holds true ; and, in the midst of our uncertainties, the 
 conjecture may be allowed to stand as more persuasive 
 than any alternative. Beyond this, Miss Hunt has 
 not much of value to contribute. She hints a vague 
 approval of Fleay's attribution of scenes 1-9 to Web- 
 ster, 11-17 to Dekker. But she qualifies this by giving 
 Dekker parts of 7 and 9, and probably 4, and Web- 
 ster 10. The pathos of the trial-scene (16), she thinks, 
 points to Dekker. 
 
 Her judgment is not very trustworthy. It is on 
 emotional rather than aesthetic grounds — she attri- 
 butes, I mean, a tender scene to Dekker and a gloomy 
 scene to Webster, because Dekker is a tender, and 
 Webster a gloomy, dramatist. 
 
 Welcoming a suggestion of Dr. Greg's, she finds the 
 speeches of Wyatt in 6 and 10 very un-Dekkerish, and 
 therefore gives these scenes to Webster. (Mr. Pierce, 
 more "scientifically" notices the same thing.) For 
 myself, speaking with all due mistrust of human ability 
 to pick out one author from another in these cases, I 
 thought I too found a different note in these scenes. 
 But if it is not Dekker's, it is as certainly neither the 
 Webster's of 1612 nor the "Webster's" of the fancied 
 
220 JOHN WEBSTER 
 
 Websterian parts of this play. It seems to me far 
 more probably Heywood.^ 
 
 The whole position is this, Sir Thomas Wyatt con- 
 sists of the fragments of the first or of both of two 
 plays, one by Chettle, Dekker, Heywood, Smith, and 
 Webster, the other certainly by Dekker, and probably 
 by the others as well. It is issued as by Webster and 
 Dekker — either because they originally had the larger 
 share, or because they did the editing, or because their 
 names were at the moment the more likely to secure a 
 sale, or because they were known as the authors of the 
 play to the publisher. In any case, it was not the cus- 
 tom to put more than two names to a play. On the 
 whole, therefore, one must begin with an a priori prob- 
 ability that most of the play as we have it is by Web- 
 ster and Dekker, but that some is by Heywood or 
 Smith or Chettle. In addition, the state of the play 
 (the text is very uneven, sometimes fairly good, some- 
 times terribly mangled), and its history of slashing and 
 patching, make it likely that the different contributions 
 are fairly well mixed together by now. In some places, 
 certainly, a delicate reader will fancy he detects re- 
 peated swift changes between more than two styles.^ 
 
 It is obvious, then, that it is very presumptuous 
 to assign different portions of the play with any com- 
 pleteness to the different authors. Reading the play, 
 with careful attention to style and atmosphere, I have 
 
 * Note especially the word "ostend," p. 194. 
 
 *e. g. the change towards the end of scene 11, at the top of 
 page 196, after Suffolk's entry. 
 
APPENDICES 221 
 
 seemed to myself to recognise in the bulk of two scenes 
 and in one or two scattered places (e.g. the opening 
 lines of the play) a voice that may well be that of the 
 younger Webster. Taking, therefore, cautiously a cer- 
 tain amount of positive evidence from Dr. Stoll and 
 Mr. Pierce, and comparing it with my own impression 
 of the play and the general impression of other critics, 
 I suggest the following conclusions as all that we can 
 fairly pretend to be more than amiable dreaming. 
 Webster probably wrote scene 2 and most of scene 16. 
 No doubt he poured indistinguishably forth other parts 
 of this commonplace bit of journalism; but, except one 
 or two lines, it is impossible to pick them out. A good 
 deal of the rest of the play is by Dekker. Heywood's 
 hand is occasionally to be suspected. 
 
Appendix D. — "Westward Ho" and 
 "Northward Ho" 
 
 These plays are so closely connected, and evidence 
 about either reacts so much on the other, that it is 
 convenient to consider them together. 
 
 Dates. 
 
 These plays can be dated fairly closely. 
 
 Westward Ho was registered to print on March 
 2nd, 1605. It was printed in 1607. 
 
 Northward Ho was registered on August 6th, 1607, 
 and printed in that year. 
 
 Northward Ho contains an amiable farcical attack 
 on Chapman.^ For this reason and others, it must 
 have been written as an answer to Eastward Ho, which 
 was registered to print September 4th, 1605, and ap- 
 peared in several editions in that year, and was prob- 
 ably written in 1604, perhaps in 1605.^ Eastward Ho 
 was written, again, more or less in emulous succession 
 
 *This is fairly conclusively proved by Dr. StoU (pp. 65-69). The 
 only doubtful point is that Bellaraont (whom we suppose to mean 
 Chapman) is called "white" and "hoary." Chapman was only 
 forty-seven in 1606. But even in this age, when people live so much 
 more slowly, they are sometimes silver-haired before fifty. And 
 the other evidence is very strong. 
 
 'v. Eastward Hoe, ed. F. E. Shelling. Belles Lettres Series, 
 Introduction. 
 
 222 
 
APPENDICES 223 
 
 to Westward Ho} So we have the order of the plays 
 fairly certain. Dekker and Webster wrote theirs for 
 the Children of Paul's; Eastward Ho was written for 
 the rival company, the children of the Queen's Revels, 
 by Chapman, with the help of Jonson and Marston. 
 
 Westward Ho, therefore, could have been written 
 any time before March, 1605. The probable date of 
 Eastward Ho makes it slightly desirable to put the 
 performance of Westward Ho back, at least, towards 
 the beginning of 1604. There are various references; 
 to Kemp's London to Norwich Dance (1600) ;2 per- 
 haps to James' Scotch Knights ; ^ and to the famous 
 siege of Ostend.* Ostend was taken in September, 
 1604, and the second quotation, at least, looks as if 
 it was written after that. It may, however, have been 
 written during the last part of the siege. And these 
 references may, of course, not be of the same date as 
 the rest of the play. But it seems fairly safe to date 
 it as 1603 ^ or 1604, with a slight preference for the 
 autumn of 1604.^ 
 
 * V. Eastward Ho. Prologue. ' Westward Ho, p. 237. 
 
 » Westward Ho. pp. 217, 326. * Westward Ho, pp. 210, 235. 
 
 ''The end of 1603, of course. All the summer the plague was 
 raging. 
 
 "a. Dr. StoU (p. 63) finds in the Earl's discovery {Westward 
 Ho, 233), of a hideous hag in the masked- figure he had thought 
 a beautiful woman, a possible reminiscence of Marston's Sopho- 
 nisba, which may have been on the stage in 1603 or 1604. But the 
 idea is a common enough one in all literatures. And if there is 
 a debt, it might almost as easily be the other way. In any case, the 
 date is not influenced. 
 
 b. If the autumn of 1604, then, of course, Eastward Ho must be 
 put on to 1605. 
 
224 JOHN WEBSTER 
 
 Northward Ho, then, must have been written in 1605, 
 1606, or 1607. In Day's The Isle of Gulls (printed 
 1606) there seems to be a reference to these three 
 plays, ^ in a passage that must have been written for 
 a first performance; which cuts out, at least, 1607, 
 and the last part of 1606. Dr. Stoll records also ^ a 
 close parallel with a passage in Marston's The Fawn, 
 He thinks The Fawn is the originator, and that it was 
 written in 1606.^ But he dates it by a very uncertain 
 reference to an execution. It is generally dated earlier, 
 and Marston may have imitated Northward Ho, or the 
 passages may, as in another Marston-Webster case, 
 have been taken independently otherwhence. So the 
 safest date for Northward Ho is 1605.* 
 
 Sources, 
 
 Westward Ho and Northward Ho are ordinary 
 citizen-comedies. The sources of these are generally 
 unknown. The plots were probably invented or 
 adapted from some current event or anecdote. As 
 Mr. Arnold Bennett says (thinking of such bourgeois 
 subjects as these plays deal with), there is no difficulty 
 about a plot; you can get a plot any time by going 
 into the nearest bar and getting into conversation over 
 
 ^Ed. Bullen: pp. 5, 6. The reference is the more probable that 
 The Isle of Gulls was written for the same company as Eastward 
 Ho. 
 
 ' P. 16. " Stoll, p. 17. 
 
 *Miss Hunt (Thomas Dekker, pp. 101-103) comes to much the 
 same conclusion; i.e. Westward Ho, 1604, Eastward Ho, 1604-5, 
 Northward Ho, 1605, as probable. 
 
APPENDICES 225 
 
 a drink. The Elizabethans, no doubt, did this. All 
 that was wanted was some intrigue on the old citizen's- 
 wives-gallants theme that would allow of practical jok- 
 ing, bawdy talk, and a little broad conventional char- 
 acter-drawing. Dr. Stoll ^ and Mr. Pierce ^ have 
 pointed out that various incidents in these plays have 
 similarities in other plays of Dekker's earlier or later. 
 The "borrowing" from Sophonisba I have dealt with. 
 The ring story in Northward Ho is paralleled in Male- 
 spini's Ducento Novelle,^ as Dr. Stoll points out. It 
 can be traced further back (to the detriment of Dr. 
 StoU's suggestion that it originated in an exploit of 
 some attendants on Cardinal Wolsey), to number sixty- 
 two in La Sale's Les Cent nouvelles Nouvelles, a collec- 
 tion of the middle of the fifteenth century.* From 
 La Sale it could easily have come into any of the 
 Elizabethan books of stories, directly or by degrees. 
 Or it might even have been merely reinvented. 
 
 Collaboration, 
 
 Dr. Stoll has given some pages, and Mr. Pierce two- 
 thirds of his book, to an elaborate attempt to divide 
 up these plays between Dekker and Webster. It is 
 not possible here to examine either their methods or 
 their results in detail. I can only suggest some prin- 
 ciples which should be kept in mind in attempting such 
 questions, and which they have not always kept in mind, 
 
 *Pp. 72-74. 
 
 ^ The Collaboration of Webster and Dekker, Chap. VI. 
 
 ' Novella II., not I., as Dr. Stoll gives it. 
 
 *v. Celio Malespini und seine Novellen: Misteli. 
 
226 JOHN WEBSTER 
 
 and summarise their results, indicating how far they 
 seem vahd and valuable. I shall mostly consider Mr. 
 Pierce's work, as it is later and far more detailed than 
 Dr. Stoll's and includes it.^ 
 
 Dr. Stoll finds that the general outline and spirit 
 of the plays, the characters, and most of the incidents 
 are repeated in Dekker's other city-plays. On these 
 grounds, and on grounds of style and phrase, he gives 
 Dekker, in a general way, the whole of the plays. Mr. 
 Pierce adopts a more systematic method. He employs 
 various tests, "scientific" and "aesthetic," separately, 
 and tabulates and compares the results. His tests are 
 of the following kinds ; parallel passages ; use of dia- 
 lect ; metrical ; incidents ; "character and atmosphere" ; 
 and the "three-syllable Latin-word test," an invention 
 of his own. The last needs explanation. Mr. Pierce 
 discovered that the difference in typical passages of 
 Webster and Dekker, the difference of weight and 
 rhythm, is partly due to the number of long Latin 
 words used by the former. He has made this into a 
 regular and usable test, by reducing all Webster's and 
 Dekker's plays to a common line measure, and finding 
 the percentage of three-syllable words of Latin or 
 Greek origin, in each scene and act. An ingenious 
 plan. The results are superficially of immense decision 
 and value. Webster's known plays have a high aver- 
 age ; Dekker's known plays a low one. A few scenes in 
 these two collaborate plays have a high average, and 
 
 * See also a very sensible review of Mr. Pierce's book by Dr. P. 
 Aronstein in Beiblatt zur Anglia, 1910, p. 79. 
 
APPENDICES 227 
 
 the rest a low one. There is a wide, almost empty gap 
 in between. The conclusion, especially if other tests 
 agree, is obvious. 
 
 But this test makes certain assumptions which Mr. 
 Pierce does not seem to have considered. It assumes 
 that the use of these three-syllable Latin words is al- 
 ways independent of the subject-matter. It assumes 
 that it was, even at this date, not only a habit of Web- 
 ster's, but an ingrained one, and probably unconscious. 
 If (and it is very probable) he was merely forming his 
 style at this time, by imitating such writers as Marston, 
 he could and would drop this trick a good deal, or for- 
 get to keep it up, in writing this sort of play. Writers 
 are not born polysyllabic. The habit may supremely 
 suit them; but they acquire it. And the process of 
 acquiring it is generally conscious. When Webster 
 wrote (or copied out) 
 
 "I remember nothing. 
 
 There's nothing of so infinite vexation 
 
 As man's own thoughts." 
 or 
 
 "I have caught 
 
 An everlasting cold: I have lost my voice 
 
 Most irrecoverably." 
 
 he knew what he was doing as well as Mr. Henry James 
 does when he writes, "She just charmingly hunched her 
 eyes at him." 
 
 If the investigators of the future draw up lists of 
 the average number of adverbs to a uniform line in 
 Mr. Henry James' works, they will find, probably, that 
 
228 JOHN WEBSTER 
 
 in the early works it is practically normal, in the early- 
 middle period uneven, varying from chapter to chap- 
 ter, and for the last twenty years immense. Who they 
 will think wrote the early, and collaborated in the mid- 
 dle, Henry James's, it is impossible to guess. 
 
 That this Latinism could be put on at will we have 
 Dekker's The GulVs Horn-Booh and passages in his 
 more serious plays to witness. In spite of that it may 
 be admitted that a quite high average in any scene in 
 Northward Ho or Westward Ho, where Dekker would 
 have no temptation to Latinise, does point to Webster. 
 But what Mr. Pierce does not seem to realise is that 
 a low average does not point in the same way to Dekker. 
 For as there is no play of this kind by Webster extant, 
 it is impossible to say how much he might have de- 
 scended from Latinity at times. It is all part of the 
 general error of taking, as Webster's normal usages, his 
 practices in a definite kind of play in his mature period. 
 Still, with these restrictions and in this way, Mr. 
 Pierce's Latin-word test has a good deal of value ; that 
 is to say, for deciding what is Webster's, not what is 
 not. The only thing that can be urged against it is 
 that it is unnecessary ; being only a symptom of a dif- 
 ference in style which a subtle taste should distinguish 
 on its own qualities, or, if more, misleading. This is 
 mostly true; and the aesthetic tests are ultimately the 
 most valuable. But then it is so hard either to fix or 
 to communicate them. 
 
 The tests of metre, incident, and character and at- 
 mosphere seem to me to have practically no value, ex- 
 
APPENDICES 229 
 
 cept in so far as "atmosphere" means literary style. 
 What it mainly means is the complexion of the whole, 
 with regard to which Westward Ho is of course much 
 nearer to, say, TJie Honest Whore, than it is to The 
 Duchess of Malfi. No doubt there are minor, barely 
 visible, effects and individualities of metre, phrase, or 
 character-drawing, and turns of incident, which might 
 easily betray the Dekker of this period, whom we 
 know, or even the Webster, whom we fear we mightn't 
 recognise. Dr. Stoll, indeed, has used these a little, for 
 distinguishing Webster. But as a rule these details 
 are just those one cannot tabulate. The grosser ones, 
 that can be defined and listed, are the attributes of the 
 species of play, such as a dramatist can put on and 
 off at will. The subtler, less extricable peculiarities, 
 however, are what influence the "unscientific" critical 
 taste to feel, "This is Webster!" and "This Dekker!" 
 They have an ultimate voice in deciding attributions, 
 though by a diff*erent method from metrical or word- 
 tests ; by representation rather than plebiscite. 
 
 The second trustworthy kind of evidence, then, for 
 a passage or scene being by some author, is a percep- 
 tion that the literary and linguistic style is his. To 
 use this, which Swinburne called judging by the ear in- 
 stead of the fingers, is a very important method, if 
 not so supreme as he thought. It is without rules ; but 
 in this case there are certain general features of style 
 which can be mentioned, if not tabulated. For Dekker 
 there is the half-comical, quick, repetition of phrases, 
 that Dr. Stoll has noticed. There is an important un- 
 
230 JOHN WEBSTER 
 
 observed characteristic of Webster's, which is ex- 
 tremely noticeable in his later works, and seems to ap- 
 pear in those portions of these plays which, on stylistic 
 and other grounds, we are led to believe his. It is in 
 marked contrast to Dekker. It is the use of involved 
 sentences with subordinate clauses, as against a style 
 where the ideas are expressed in a series of simpler, 
 shorter, co-ordinate sentences. Northward Ho, II. 2, 
 one of the only certainly Websterian scenes in the two 
 plays, strikes the ear immediately as different in this 
 way. The whole ring of the sentences is — mainly for 
 this reason — slower, deeper, more solemn. The Ger- 
 mans have invented a way of distinguishing collabora- 
 tors. Read the play, they say, and you find your 
 voice instinctively assumes a different pitch for the 
 work of different authors. They profess to tell to half 
 a sentence where Webster begins and Dekker leaves off. 
 One can smile at their whole claim. But, for these two 
 authors, it is not, essentially, unmeaning. 
 
 The third admissible way of dividing the authorship 
 of these plays, is by parallel passages. It is not gen- 
 erally kept in mind that if this method is used for de- 
 ciding between collaborators, it implies an assumption 
 that the collaboration was of a certain kind, namely, 
 by taking so many scenes each. This was the usual 
 practice in contemporary collaboration, we know; and 
 it is, obviously, far the quickest and easiest way, as a 
 rule. So we have a right, generally, to suppose that 
 collaboration was of this sort, and, therefore, that a 
 certain parallel or repetition is strong proof of au- 
 
APPENDICES 231 
 
 thorship of that scene. All the same, there is always 
 the possibility of both authors working over the same 
 scene, in which case, of course, a parallel helps to prove 
 nothing except its own source. In the present case, 
 though we do not know so certainly as with Webster's 
 earlier plays, Sir Thomas Wyatt or Christmas comes 
 hut once a year, that the collaboration was real and 
 contemporary, it is very likely. The likehhood is made 
 smaller than usual by the facts that Dekker was a 
 much quicker worker than Webster, and that he was 
 by standing and experience the senior partner. He 
 might very well have gone over Webster's scenes. 
 
 On the whole then a single parallel or repetition 
 does not prove much, in these plays ; a row of them, in 
 one scene, goes far to establish the authorship of that 
 scene. 
 
 Mr. Pierce has collected a great number of possible 
 parallels, most of them insignificant, some of them 
 very valuable. In using them, one must remember that 
 we have only a very few, and quite different, later plays 
 by Webster to draw on, and a great many, some con- 
 temporary and similar, of Dekker's. Once again, ab- 
 sence of proof that a scene is Webster's does not prove 
 it is not. 
 
 By these methods of proof, and any outstanding 
 evidence of another kind, one reaches much the same 
 conclusions as Mr. Pierce; but, I think, they should 
 be applied differently. In Northward Ho, II. 2, and 
 the first part of V., are almost certainly in the main by 
 Webster. In Westward Ho there is not, it seems to 
 
232 JOHN WEBSTER 
 
 me, the same certainty. But I. 1 and III. 3 show very 
 strong traces of his presence. With Northward Ho, 
 
 1. 1 and III. 1 the probability is smaller, but still con- 
 siderable. There are also one or two phrases or sen- 
 tences scattered about the plays that arrest one's at- 
 tention as recognisably Webster's, or at least not Dek- 
 ker's. But these do not extend their atmosphere be- 
 yond themselves. There are these few scenes, which, 
 with varying degrees of probability, can be given to 
 Webster. There are a few more {Westward Ho, II. 1, 
 
 2, V. 3 : Northward Ho, IV. 1 ) where all the evidence 
 points to Dekker being mainly responsible. In the 
 rest, while we cannot detect the Webster of 1612, we 
 have no right to deny the presence of the Webster of 
 1605. In any case the collaboration seems to have been 
 of an intricate and over-laid nature. 
 
 To pretend to more precise knowledge is, I think, 
 silly. 
 
 Since I wrote this. Miss Hunt's book on Thomas 
 Dekker has appeared. On pages 106, 107, and 108 she 
 discusses the shares of Webster and Dekker in these 
 plays. She principally follows Fleay, whose methods 
 were rough. She discusses the responsibility for the 
 plots, which other critics have been inclined to leave 
 vaguely to Dekker. She would give most of it to Web- 
 ster, and also "the more unusual subtle or abnormal 
 incidents" ; the device of the diamond in Westward Ho 
 and that of the ring in Northward Ho, perhaps also 
 Greeneshield's betrayal of his wife, although that may 
 have been borrowed from Eastward Ho, Also Jus- 
 
APPENDICES 233 
 
 tiniano's disguise as a hag; and his and Mayberry's 
 jealousy. Other kinds of evidence she does not con- 
 sider. In Westward Ho she finds signs of incomplete 
 collaboration and change of plan in construction. Still 
 following Fleay she thinks Webster wrote most of Acts 
 L, II., and III., and some of IV. ; Dekker, the rest. 
 Northward Ho is more homogeneous. Dekker is given 
 the Chapman-ragging and the Doll scenes ; Webster the 
 rest. Dekker probably went over the whole. 
 
 Her proofs and judgments are very superficial, and 
 almost valueless. It is, perhaps, probable that Web- 
 ster had more share in the planning of the plots and 
 incidents than he has been allowed. Her assignments 
 in general are based on a feeling that these two plays 
 are "gross," "offensive," and "sinning against the 
 light," that her protege Dekker, being a pure-minded 
 man, can have had little to do with them, and that 
 Webster "who dealt with lust" must be held guilty. 
 Her sex, or her nationality, or both, have caused in 
 her a curious agitation of mind whenever she ap- 
 proaches these plays. This prejudice destroys what 
 little value her very cursory investigation of the prob- 
 lems of their authorship might otherwise have had. 
 
Appendix E. — "The Malcontent" 
 
 The Malcontent was published in 1604, in two edi- 
 tions. The title-page of the first reads : 
 
 THE 
 MALCONTENT. 
 
 BY JOHN MARSTON. 
 
 The title-page of the second reads: 
 
 THE 
 MALCONTENT. 
 
 AUGMENTED BY MARSTON. 
 
 WITH THE ADDITIONS PLATED BY THE KIKGS 
 MAJESTIES SERVANTS. 
 
 WaiTTEN BY JOHN" WEBSTER. 
 
 The second edition differs from the first in having an 
 Induction, and the insertion of twelve passages in the 
 play. 
 
 Much fuss has been made about the amount of the 
 play that Webster wrote. Dr. Stoll ^ has conclusively 
 shown that all we can deduce to be Webster's is the 
 *Pp. 55-60, 
 
 234 
 
APPENDICES 235 
 
 Induction ; and Professor Vaughan has called attention 
 to a final piece of evidence — that the Induction itself 
 practically says that this is the case. 
 
 The matter is quite clear. The full-stop after 
 "Servants" on the second title-page is what Dr. Stoll 
 calls "purely inscriptional." That the whole theory 
 of Elizabethan punctuation rests on a psychological, 
 not, as now, on a logical basis, has recently been shown 
 with great force by Mr. Simpson.^ The whole look of 
 the page makes it obvious that the intention was to 
 connect Webster with the "Additions," and only with 
 the additions, and to make Marston responsible for the 
 augmentations as well as the bulk of the play. An aes- 
 thetic judgment of the play declares that the extra 
 passages are all Marston's and that the Induction is 
 probably not by Marston and probably is by Webster. 
 And Burbadge, in the Induction, describing how the 
 play fell into the hands of the King's Servants (from 
 the Children of the Queen's Revels) and being asked 
 "What are your additions?" makes answer, "Sooth, 
 not greatly needful; only as your salad to your great 
 feast, to entertain a little more time, and to abridge 
 the not-received custom of music in our theatre." That 
 probably, though not quite necessarily, identifies the 
 "additions" with the Induction. There are three pos- 
 sible theories; that Marston wrote The Malcontent 
 (first edition) and the extra passages, and Webster 
 
 ^Shakespearian Punctuation. See also Professor Grierson's re- 
 marks on Elizabethan punctuation, The Poems of John Donne, vol. 
 ii., pp. cxxi.-cxxiv. 
 
236 JOHN WEBSTER 
 
 the Induction; that Marston wrote The Malcontent 
 (first edition) and Webster the extra passages, and 
 probably the Induction ; or that originally Marston 
 and Webster wrote the play together, and that for 
 some reason only Marston's name appeared on the 
 title-page. I think there is no reason to believe the 
 third, every reason not to believe the second, and sev- 
 eral reasons to believe the first. I do not think the 
 arguments for The Malcontent dating from 1600, and 
 for the "augmentations" being really restorations by 
 Marston of cut pieces of his play in its first state, are 
 decisive. But I think the case stands without these 
 conclusions.^ 
 
 Date, 
 
 As the first edition appeared without the Induction 
 during 1604, and the second with it in the same year, 
 and as it was obviously written for a special piratical 
 revival by the King's Majesty's Servants, who claim 
 the second edition, it is fair to suppose that the In- 
 duction was written during 1604. 
 
 * On the date of The Malcontent Dr. Stoll goes off pursuing the 
 wildest of geese through the undergrowth of a footnote. He 
 "proves" a phrase to be in the "Ur-Hamlet" by taking it for 
 granted that a play printed in 1604 is exactly as it was when it 
 was written in 1600. The old assumption of the integrity of plays. , 
 
Appendix F. — "The White Devil" 
 
 Date. 
 
 The White DevU was printed in 1612. It obviously 
 belongs to the same period as The Duchess of Malfi. 
 That it is the earlier of the two is probable on general 
 grounds, and proved by the advance of metrical li- 
 cense ^ and the absence of phrases and adaptations 
 from the Arcadia, which are present in all Webster's 
 later work.^ 
 
 There are various clues, of more or less relevance, 
 to its date: 
 
 Mr. Percy Simpson has pointed out ^ that the puz- 
 zling and much emended passage about Perseus (p. 21 ; 
 last line) is an allusion to Jonson's Masque of Queens 
 (1609); a work Webster knew, for he borrows in A 
 Monumental Column from the dedication to it. 
 
 P. 23. MoNTiCELSo. Away with her! 
 
 Take her hence! 
 ViTTORiA. A rape! a rape! 
 
 MONTICELSO. How? 
 
 ViTTORiA. Yes, you have ravished Justice; 
 
 Forced her to do your pleasure. 
 
 * V. StoU, p. 190, metrical table. 
 
 » F. Crawford, Collectanea, i., 20-46. It is very noticeable, and 
 only to be explained by Webster having filled his notebook from 
 the Arcadia after The White Devil and before The Duchess of 
 I, A Mormmental Column, and The Devil's Law-case. 
 
 * Modern Language Review: January 1907. 
 
 237 
 
238 JOHN WEBSTER 
 
 Dr. StoU suggests that Vittoria's cry, in its sudden- 
 ness as well as in the words, is very like Sebastian's 
 in Tourneur's The Atheisfs Tragedy, I. 4. But any 
 connection between the two is doubtful ; if there is any, 
 Tourneur may have imitated Webster ; and anyhow the 
 date of The Atheisfs Tragedy is still quite uncertain 
 — 1607-1611 is the most definite limit one can venture, 
 and even that rather depends on accepting the anony- 
 mous Revenger^s Tragedy as Tourneur's. This pas- 
 sage is more likely to be connected with The Tragedy 
 of Chahoty V. 11, 122, "unto this he added a most 
 prodigious and fearful rape, a rape even upon Justice 
 itself. . . ." Professor Parrott thinks Chapman may 
 have written this (it is in his part of the play) about 
 1612. And Webster admired and imitated Chapman. 
 But the whole thing is too cloudy for the resemblance 
 to be more than interesting. 
 
 The number of references to Ireland in the play is 
 remarkable.^ Either Webster had been in Ireland, or 
 he had been hearing about it, or he had been reading a 
 book on it. If it was a book, Barnaby Rich's A New 
 Description of Ireland, 1610, has been suggested. It 
 is very probable ; for the book mentions the various sub- 
 jects of Webster's references. But as there is no ver- 
 bal connection, and as they are all things one could 
 easily pick up by hearsay, the proof is not conclusive. 
 No doubt, too, there were other books on Ireland at 
 
 *See p. 6. Irish gamesters: p. 16, no snakes in Ireland: p. 28, 
 Irish rebels selling heads: p. 29 "like the wild Irish. . . .": p 31, 
 Irish funerals. 
 
APPENDICES 239 
 
 the time which might have contained such obvious jour- 
 nahstic prattle as this. Still, Rich's book is the best 
 explanation of Webster's mind being so full of Irish 
 facts at the time: and the references are scattered 
 enough to make a little against them having been intro- 
 duced in a revision. For what this sort of evidence is 
 worth, it points to 1610 or after. 
 
 Dr. Stoll attaches importance to the preface and 
 postscript. These, it would in any case be extremely 
 probable, were written in 1612 for the publication 
 of the book. And a pretty conclusive borrowing of 
 phrase from Jonson's preface to Cat aline (1611)^ con- 
 firms this. Dr. Stoll thinks the tone of the preface 
 shows that the performance was recent. It is difficult 
 to see why. Webster merely says that the play has 
 been performed, without much success. His only hint 
 about the time that has elapsed since lies in "and that, 
 since that time [^i.e. the time of the performance], I 
 have noted most of the people that come to that play- 
 house resemble those ignorant asses, who, visiting sta- 
 tioners' shops, their use is not to inquire for good 
 books but new books. . . ." This looks as if some time 
 had gone by between the performance and the writing 
 of the preface. He had had time to see and deplore 
 The White Devil being forgotten by the "ignorant 
 asses" who only wanted "new" goods. An interval of 
 some months should be allowed at least. 
 
 The preface gives the further information that the 
 
 *See Stoll, pp. 20, 21. Webster borrows most of this preface 
 from prefaces of Jonson and Dekker. 
 
240 JOHN WEBSTER 
 
 performance had been in winter, and that the play had 
 taken a long time in writing. 
 
 There is one more point. Dekker, writing an Epistle 
 Dedicatory to // This be not a Good Play ^ addressed 
 to the Queen's Servants (who produced The White 
 Devil), wishes well to a new play by a "worthy friend" 
 of his. It has been suggested that this means The 
 White Devil. Dekker and Webster were old friends, 
 and the vague complimentary epithets of the play 
 apply. ^ It may be so. But as between twenty and 
 thirty new plays were produced every year,^ and the 
 Queen's Servants, no doubt, contributed their share, 
 there were a good many other plays Dekker might have 
 been thinking of, and we cannot regard this as more 
 than a possible conjecture. // This be not a Good Play 
 was probably written and played in 1610 or 1611. The 
 Epistle Dedicatory for the printed edition would prob- 
 ably be written for the occasion, i.e. in 1612 or the 
 end of 1611. So any weight this conjecture has would 
 point to Webster's play being produced in the begin- 
 ning of 1612.* 
 
 * Printed 1612. 
 
 '"Such brave Triumphs of Poesy and elaborate industry . . ." 
 ' V. Schelling, Elizabethan Drama, ii. pp. 371, 373. Malone and 
 
 Fleay both suggest an average of twenty-three or four a year. 
 
 This period was more prolific than the average, of course. For 
 
 1601-1611 Professor Schelling surmises a yearly average of nearer 
 
 thirty. 
 
 * Dr. StoU offers the additional proof that Dekker is speaking of 
 a maiden effort, which The White Devil is. Mere assumptions. 
 Dekker does not say the object of his interest is a maiden work. 
 And nobody can state that The White Devil is. 
 
APPENDICES 241 
 
 The similarity of style and atmosphere and the close 
 resemblance of a great many passages^ (not verbal 
 repetitions, far more subtle and convincing things than 
 that) make it desirable to put The White Devil and 
 The Duchess of Malfl as close together as possible. 
 The tenuous evidence we have noticed points, if any- 
 where at all, to agreement with this — that is, to put- 
 ting The White Devil on towards its final hmit of 1612. 
 Acknowledging that it is all quite uncertain, I think 
 it is most probable that the play was written during 
 1611 and performed at the end of that year or in 
 January or February, 1612. It may have been writ- 
 ten 1610 and performed 1610-1611. It would need 
 some strong new evidence to put it back further. 
 
 Sources. 
 
 Some time and trouble have been spent in seeking 
 an exact printed source for The White Devil, but, so 
 far, in vain. The actual events, which took place in 
 the end of the sixteenth century — Vittoria was born 
 in 1557, was murdered in 1585 — were well-known.^ 
 Did Webster get the story from an accurate history, 
 from some romantic version, or from hearsay? One 
 can only surmise. Professor Vaughan, who goes at 
 greatest length into this question, thinks it quite pos- 
 sible the source was a novel or play, or an oral account, 
 
 ^See, for examples, Sampson, Introduction to The White Devil, 
 etc., pp. xli.-xliii. and Stoll, pp. 80-82. 
 
 ^ For detailed accounts see D. Gnoli, Vittoria Accoramhoni. J. 
 A. Symonds, in Italian By-ways (1883) : L. M'Cracken, A Page of 
 Forgotten History. 
 
242 JOHN WEBSTER 
 
 but is most in favour of Webster having read some 
 fairly accurate contemporary account, and altered it 
 for dramatic purposes. Webster's unusually accurate 
 pronunciation of Italian names, and his quoting Tasso,^ 
 allow us to believe he may have known Italian. But the 
 tale may well have got into an English or French ver- 
 sion by 1610. The differences between Webster's ver- 
 sion and the facts are queer. Many of them look cer- 
 tainly as if they had been made consciously (by Web- 
 ster or someone else) for dramatic purposes; such as — 
 besides the additions of madness and murders — the 
 toning down of Lodovico to make him a minor figure, 
 and the purification of Isabella. But there are others 
 that have no such obvious point, the exchange of names 
 between Marcello and Flamineo, the writing of Monti- 
 celso for Montalto,^ and Paul IV. for Sixtus V. The 
 first of these may be purposeful. Even one who has 
 not read the Sixth ^neid may be able to perceive that 
 Marcello is a pure young hero and Flamineo an amaz- 
 ing villain. Is it fanciful to more than suspect that 
 The White Devil would be less effective if he were called 
 Flamineo who died so innocently, and a Marcello playeii 
 amazing tricks with bulletless pistols, or screamed in 
 mock-death : 
 
 * The Duchess of Malf,, p. 78. 
 
 *Dr. Greg {Modern Language Quarterly: Dec. 1900) suggests 
 that Webster may have misread (in, perhaps, a MSS. account) 
 Moncelto for Montalto, and euphonised it into Monticelso. But 
 the other difficulties remain. 
 
APPENDICES 243 
 
 "O I smell soot. 
 Most stinking soot! The chimney is a-fire! 
 My liver's parboil'd like Scotch holly-bread; 
 There's a plumber laying pipes in my guts, 
 
 it scalds !" 
 
 It is not for nothing that you dare not call a hero 
 Lord John or a villain George. And Webster, who had 
 above all things a nose for irrelevant details that inex- 
 plicably trick you, unconscious, into the tone he desires, 
 may have had a purpose in writing also Paulus for 
 Sixtus, Monticelso for Montalto. Still, it is hard to 
 think memory or report or notes did not play him false. 
 On the other hand such minute details from the 
 actual story have been preserved by Webster — names, 
 the summer-house by the Tiber, and so on — that it is 
 difficult to imagine that he got it from any scanty or 
 oral report. And there are certain consideration^ 
 which seem to favour his having worked from some ex- 
 tensive version, whether dramatic or in pamphlet form. 
 Why should Brachiano and the Conjuror conduct their 
 interview in Vittoria's house (p. 18)? No reason is 
 given for the absurdity. There is an equally unex- 
 plained and apparently pointless incident in the trial- 
 scene ; where Brachiano refuses a chair, and sits on his 
 cloak (pp. 19 and 22), to show, one gathers, his con- 
 tempt for the Court. The labour and time Webster 
 spent on the play, and his care in publishing this edi- 
 tion to wipe out the failure of the performance, forbid 
 our explaining these things by hurry in composition, or 
 by the text being printed from an acting version. They 
 
244 JOHN WEBSTER 
 
 might well be the result of Webster's obvious lack of 
 ordinary skill in dramatising a story of which he had 
 a lengthy version before him. Such incidents as Fran- 
 cisco's sight of Isabella's ghost, and the spectacular 
 and fairly accurate ceremony of choosing a Pope, as 
 well as the divergencies in the characters of Francisco 
 and Flamineo, as the play proceeds, also fit in well with 
 this theory. 
 
 If Webster was working from some detailed account, 
 it might either be a play or a narrative. In favour of 
 the play are some of the extraordinary old-fashioned 
 tags in The White Devil, and particularly the amazing 
 mixture of extremely fine and true lines and distress- 
 ingly ludicrous couplets or phrases in the final scene 
 (though such incongruities are far more possible for 
 Webster than for any other great writer of the period). 
 In this case, the characteristics of the dramatisation 
 are due to the earlier play-wright. 
 
 On the other hand, the general line of the play gives 
 the impression that Webster himself dramatised it di- 
 rectly. 
 
 In any case, from the details of names mentioned 
 above, it looks as if someone, either Webster or an 
 intermediate, had read some accurate account with 
 care, making a few notes perhaps, had let it simmer 
 into shape in his mind, the characters taking life and 
 individuality, and then, later, written it out. Only so 
 can the mistakes of memory be explained. Whether 
 it was Webster who did this, or whether, as Professor 
 
APPENDICES 245 
 
 Vaughan implies, he had someone else's account before 
 him as he worked, it is impossible to say. 
 
 The State of the Play. 
 
 The White Devil is certainly entirely Webster's. It 
 is also almost certain we have the whole play. There 
 are no sure traces of revision for acting, or of abbrevi- 
 ation. Webster obviously, from his Preface, brought 
 the play out with great self-consciousness and care, and 
 a desire to see its merits recognised. So he would 
 naturally print it complete. And both the Preface and 
 general probabilities point to it having only been played 
 once, not very successfully, before publication. So we 
 need not suspect our copy of having been revised for a 
 revival. 
 
Appendix G. — "The Duchess of Malfi" 
 
 Date. 
 
 The history of the various opinions about the date 
 of The Duchess of Malfi is both entertaining and in- 
 structive. Dyce used to guess at 1616. Fleay put it 
 back to 161S, a date which many slight indications 
 favoured. These were mainly on stylistic and general 
 grounds. Professor Vaughan, however, in 1900, made 
 a suggestion which Dr. Stoll, in 1905, worked out and 
 regarded as providing conclusive evidence. So, accord- 
 ing to the ordinary methods of dating plays, it did. 
 It is not necessary to detail Dr. Stoll's arguments. 
 They refer to the oddly introduced passage in I, i. (p. 
 59) on the French King and his court. Dr. Stoll 
 rightly says it is very probable a passage like this in an 
 Elizabethan play would refer to current events. He 
 exhaustively proves that it does exactly fit what hap- 
 pened in France in the early part of 1617, when Louis 
 XIII. had the evil counsellor Concini killed, "quitted" 
 his palace of "infamous persons," and established a 
 "most provident council" ; events which made some stir 
 in England at the time. As all this would have ap- 
 peared in a different light in 1618 or after, and as there 
 is other evidence that The Duchess of Malfi was being 
 played in England at the end of 1617, we seem to have 
 the date, the latter part of 1617, fixed with unusual 
 
 246 
 
APPENDICES 247 
 
 certainty.^ It is rare to be able to be so certain and so 
 precise about an Elizabethan play. And having the 
 date of composition of some thirty lines fixed, people 
 would no doubt have gone on for ever believing they 
 had the date of the whole fixed ; had not Dr. Wallace, 
 delving in the Record Office, discovered that William 
 Ostler, who played Antonio, died on December 16th, 
 1614 ! 2 The explanation, of course, is that The Duchess 
 of Malfl was written and performed before December, 
 1614, and revived with additions in 1617. All the evi- 
 dence we have shows that this habit of altering a play 
 and putting in topical references whenever it was re- 
 vived, was universal. Our modern reverence for the 
 exact written word is the result of regarding plays as 
 literary objects, and of our too careful antiquarian 
 view of art. The Elizabethans would have thought it 
 as absurd not to alter a play on revival as we think 
 it to do so. They healthily knew that the life of a play 
 was in its performance, and that the more you inter- 
 ested people by the performance, the better it was. The 
 written words are one kind of raw material for a per- 
 formance; not the very voice of God. So, naturally, 
 they changed the play each time ; and when we have the 
 text of a play, all we can feel in the least certain about, 
 is that we have it something as it was for the latest 
 
 *See, for instance, Professor Schelling, Elizabethan Drama, vol. 
 i, p. 590. "This fixes the date of The Duchess of Malfi at a time 
 later than April, 1617, and puts to rest once and for all former 
 surmises on the subject." This eternal rest lasted nearly five years. 
 
 » See The Times, Oct. 2 and 4, 1909. 
 
248 JOHN WEBSTER 
 
 previous revival. Editors and critics have come to ad- 
 mit this, in general. But in individual instances they 
 never remember to allow for it. Occasionally, as here, 
 other circumstances are discovered, and put them right. 
 But, on the whole, the common credulous assumption 
 of certainty about dates in Elizabethan literature is as 
 startling to an onlooker as the credulous assumption of 
 certainty about authorship. 
 
 The Duchess of Malfi, then, was acted before Decem- 
 ber, 1614; and as Webster obviously took as long over 
 it as he confessedly did over The White DevU, the latest 
 date we can give him for writing it is during the whole 
 year of 1614. As it is later than The White Devil, we 
 do not want to put it back beyond 1612, though as The 
 White DeviVs date is uncertain we could do so. 
 
 Strong internal evidence for the date of The Duchess 
 of Malfi has, however, been pointed out by Mr. Craw- 
 ford.^ His arguments rest mainly on the great sim- 
 ilarity between The Duchess of Malfi and A Monumental 
 Column. These are connected far more closely than any 
 of Webster's works in several ways. The poem repeats 
 both more words and lines and more ideas from The 
 Duchess of Malfi than from any of the other plays. In 
 metre it is, allowing for the different styles, nearer. If 
 you examine the particular sources Webster borrowed 
 from, the resemblance becomes even more obvious. In 
 The White Devil he does not borrow from Sidney's Ar- 
 cadia at all. In The DeviVs Law-Case the borrowing is 
 
 ^Collectanea, Series i. pp. 20-46, and especially Series ii. pp. 
 1-63. 
 
APPENDICES 249 
 
 faint and patchy. In The Duchess of Malfi and A Morv- 
 umenial Column the borrowing is incessant and similar, 
 and includes imitation of style. Another work both 
 pieces borrow from, and only these two pieces among 
 Webster's, is Donne's An Anatomy of the World, which 
 was published in 1612.^ There are also ^ in The Duch- 
 ess of Malfi several imitations and borrowings of phrase 
 from another book of 1612, Chapman's Petrarch's 
 Seven Penitential Psalms. But the similarity itself of 
 A Monumental Column and The Duchess of Malfi puts 
 the date of the play further on than this. A Monumen- 
 tal Column is an elegy written in memory of Prince 
 Henry, who died on November 6th, 1612. It was pub- 
 lished in 1613, with similar elegies of Tourneur's and 
 Heywood's. It appears to have been rather belated, for 
 (lines 259-268) he refers to other elegies that had al- 
 ready appeared, and adds: 
 
 "For he's a reverend subject to be penn'd 
 Only by his sweet Homer and my friend." 
 
 i.e., only Chapman should write about the dead Prince. 
 From this and from various reminiscences in A Monu- 
 mental Column^ Mr. Crawford deduces that Webster 
 must have seen Chapman's Epicedium on Prince Henry. 
 I do not think it is proved ; for the passage may only 
 mean that Chapman ought to write an elegy. In any 
 case. Chapman's poem followed the Prince's death so 
 
 ^In its entirety. Without The Second Armiversary in 1611. But 
 Webster borrows from the whole. 
 ' Crawford Collectanea, ii. 55-58. 
 
250 JOHN WEBSTER 
 
 closely (as the other elegies Webster refers to also may 
 well have done) that we cannot put A Monumental Col- 
 umn much later for this. But (lines 102-5) there is a 
 probable, though not certain, reference to Chapman's 
 The Masque of The Middle Temple performed Febru- 
 ary 15, 1613. A Monumental Column, therefore, may 
 be dated any time in the half-year December, 1612- 
 May 1613, with a slight preference for February and 
 March 1613. As The Duchess of Malfl was certainly 
 before the end of 1614, and certainly after the begin- 
 ning of 1612, and as there is so much evidence that the 
 play and the poem were being written at the same time, 
 we may date the play with fair certainty at 1613, in- 
 cluding perhaps the latter part of 1612. 
 
 There is no other evidence of any value for the date 
 of The Duchess of Malfl. It may appear that I have 
 been trying to establish the earlier limit by that method 
 I have always decried elsewhere, namely, by dating the 
 whole by the date of various passages. The answer is 
 that in the case of The Duchess of Malfl and A Monu- 
 mental Column the borrowings from other authors are 
 so numerous, so widespread, and so much part of the 
 whole play, that the likelihood of them having all been 
 introduced in revision is very small. Such a revision 
 would have to be a complete rewriting of the play. And 
 while we must allow for the possibility of revision in any 
 Elizabethan play, we cannot suppose that the writers of 
 that age took the trouble to rewrite their plays, in tone, 
 from beginning to end. 
 
APPENDICES 251 
 
 Sources, 
 
 It Is certain that Webster got the story of The Duch- 
 ess of Malfi from Painter's Palace of Pleasure, Novel 
 XXIII. Painter had it from Belleforest, who had it 
 from Bandello. A recent Italian book shews that Ban- 
 dello probably based his account on the testimony of 
 actors in the actual events, and suggests that he may 
 even have been himself one of them, the one whom we 
 know as Delio.^ It is an alluring speculation. 
 
 Beyond this, the tortures of the Duchess were sug- 
 gested, probably, by incidents in Sidney's Arcadia. The 
 same book, which gave Webster so much even in phrases 
 and sentences, may have been responsible for much in 
 the Duchess's character, and for the echo-scene (V. 3). 
 These are less certain. Mr. Crawford with greater 
 probability thinks that V. 1., the scene of Delio's and 
 Julia's suits to Pescara, was suggested from Montaigne, 
 Book 1.2 
 
 State of the Play. 
 
 I have already explained some of the reasons for 
 thinking there was a revival of The Duchess of Malfi 
 in the latter half of 1617. They are, briefly, these. 
 The first fifty lines of the play obviously refer to events 
 which happened in France in April 1617, and roused 
 immediate interest in England. They could not have 
 
 ^ Giovanna d'Aragona, Duchessa d'Amalfi, da Domenico Morel- 
 lini, 1906. V. review by W. W. Greg in Modern Language Re- 
 view, July 1907. 
 
 * Collectanea, ii. pp. 14, 15. 
 
252 JOHN WEBSTER 
 
 been written after about May 1618, when these events 
 were seen in a quite different light. Also, the chaplain 
 to the Venetian Ambassador in England has left a de- 
 scription of a play he saw in London, which is probably, 
 but not certainly. The Duchess of Malfi} He did not 
 get to London before the beginning of October 1617, 
 and he seems to have seen the play a little time before 
 the 7th February 1618. 
 
 The Actors' list in the first edition allows of a revival 
 of this date. 
 
 The Duchess of Malfi, then, was revived in a revised 
 form in the latter part of 1617. That the beginning 
 of the play was revised we know. If the Italian chap- 
 lain's account of the play be accurate, there must have 
 been a good deal in the performance he saw which is 
 not in the play as we have it — even allowing for his 
 misinterpretation. 
 
 One passage in the play itself may point to a com- 
 bination of two versions. In I. 1., (p. 61) Delio use- 
 fully questions Antonio about the other chief charac- 
 ters. Antonio gives a long description of the Cardinal ; 
 then a long description of the Duke, his brother ; then, 
 before going on to the Duchess, he reverts suddenly to 
 the Cardinal, as if he had not mentioned him, with: 
 
 **Last, for his brother there^ the Cardinal. . . ," 
 
 On the other hand, the inclusion in the first quarto 
 (1623) of Middleton, Rowley, and Ford's commenda- 
 
 » F. StoU, p. 29. 
 
APPENDICES 253 
 
 tory verses, and of Webster's dedicatory letter, as well 
 as, and more forcibly than, the avowal of the title-page,^ 
 go to show that this edition of the play is as Webster 
 would have had it. It must, therefore, be fairly near 
 the original version (1613); containing most of that, 
 with whatever of subsequent additions or changes Web- 
 ster supposed improvements. And we cannot doubt 
 that practically all of the play, as we have it, is by 
 Webster. 
 
 *"The perfect and exact Copy, with divers things printed, that 
 the length of the play would not bear in presentment." 
 
Appendix H. — "A Monumental Column" 
 
 Date. 
 
 The question of the date of A Monumental Column 
 is discussed in Appendix G in connection with The 
 Duchess of Malfi. It must have been written within 
 some six months after November 1612 ; probably about 
 March 1613. 
 
 Sources. 
 
 There is, of course, no special source for a poem like 
 this. It repeats the usual thoughts in elegies of its 
 kind ; and borrows largely in expressions and in general 
 style from Donne ; also from Sidney, Chapman, and Ben 
 Jonson. 
 
 254 
 
Appendix I.— "The Devil's Law-Case" 
 
 Date, 
 
 The DeviVs Law-Case was published in 1623. There 
 is little evidence to decide the date of its writing. 
 
 (1) There is a reference (IV. 2) to an affray in the 
 East Indies : 
 
 "How! go to the East Indies! and so many Hollanders 
 gone to fetch sauce for their pickled herrings! 
 Some have been peppered there too lately." 
 
 This almost certainly refers to a Dutch attack in 
 August 1619 on some English ships engaged in loading 
 pepper. News seems to have taken from nine to fifteen 
 months to travel between England and the East Indies. 
 London might learn, then, of this pepper business any 
 time in the latter half of 1620. The word "lately," 
 and still more the comparative unimportance and tran- 
 sience of the event, suggest that the form of the play in 
 which this sentence occurred was being acted towards 
 the end of 1620 or in the first half of 1621. If that 
 form was the only form, we cannot tell ; and we have no 
 right to assume it. The whole of the reference to the 
 East Indies is comprised in a few sentences in this one 
 place. It is entirely unnecessary to the plot, and it 
 could easily have been inserted at a moment's notice. 
 
 255 
 
256 JOHN WEBSTER 
 
 (2) It is said that the chief idea in the play, Leo- 
 nora's attempt to bastardise her son by confessing a 
 long-past adultery that as a matter of fact never took 
 place, resembles stories in the pseudo-Marlovian Lusfs 
 Dominion, The Spanish Curate, by Fletcher and Mas- 
 singer, and The Fair Maid of the Inn, by Massinger 
 and another. The Fair Maid of the Inn was probably 
 not written before 1624. The Spanish Curate was writ- 
 ten between March and October 1622. It is only just 
 possible that The DeviVs Law-Case can have been writ- 
 ten after it.^ Gerardo the Unfortunate Spaniard, an 
 English translation from the Spanish, which appeared 
 in March 1622 and was the source of The Spanish 
 Curate, may also have suggested this part of The 
 DeviTs Law-Case. But resemblances are tricky things. 
 This one, closely examined, turns out to depend largely 
 on having the confession of a past misdemeanour at a 
 public trial. And to bring in a public trial is exactly 
 the thing that would independently occur to the mind of 
 a dramatist of circa 1620, if he imagined or heard of 
 the rest of the story. The only resemblance that really 
 may mean anything is to Lusfs Dominion, where a 
 widow has a grudge against her son, because of a man 
 she is in love with. So, to defame him and deprive him 
 of the inheritance, she invents, with details, and publicly 
 confesses, a story which makes him a bastard. The 
 motives and feelings of the characters in this play cor- 
 respond far more than in those others, to The DeviTs 
 Law-Case situation. It is true Lusfs Dominion is an 
 
 » V. StoU, p. 32. 
 
APPENDICES 257 
 
 old play of 1590. But it may have been revived and 
 revised many times. Perhaps it "suggested" the idea 
 of The DeviVs Law-Case — in any of the million ways, 
 direct and indirect, in which, in real life, ideas are sug- 
 gested. But the truth is that, unless a very certain 
 source is known, the search for the suggestion of so 
 unexotic an idea as this becomes rather foolish. A half- 
 remembered story, a friend's anecdote, an inspiration — 
 anything may be responsible for any proportion of it. 
 It may be useful to trace John Keats' hippocrene ; 
 not his porridge. 
 
 (3)^ The title-page says that the play was "ap- 
 provedly well acted by Her Majesty's Servants." This 
 company, which also performed The White Devil, was 
 called by this name until March 1619, when Queen 
 Anne died. It appears to have gone gradually to pieces 
 after that. Thomas Heywood, for instance, seems to 
 have left it by 1622. In July 1622, it was recon- 
 structed, with children as well as adults, as "The 
 Players of the Revels." It probably broke up in the 
 next year. The point is, under what name did it go 
 between 1619 and 1622.? Under the old one of "Her 
 Majesty's Servants," thinks Dr. Stoll. Mr. Murray, 
 the latest investigator of the history of the Dramatic 
 Companies, says it was called by the name of "The Red 
 Bull," its theatre. What evidence there is seems to in- 
 dicate this. The corresponding (or same) company on 
 tour was generally known as "The late Queen Anne's 
 
 * For this paragraph v. English Dramatic Companies, 1558-1642, 
 by John Tucker Murray: esp. vol. i. pp. 193-200. 
 
258 JOHN WEBSTER 
 
 players." We should have expected one of these two 
 latter names, if the play had been performed only be- 
 tween 1619 and 1622. This consideration by itself 
 makes a slight, a quite slight, probability of the play 
 being acted before March 1619. 
 
 Altogether, therefore, we can only say that the play 
 is earlier than July 1622, and was almost certainly 
 being acted in some form in about August 1620-July 
 1621. Everything else is quite uncertain; except that 
 the nature of the play forbids you to look earlier than, 
 at earliest, 1610. The tiny probability of 1620 or 
 after, for the whole play, established by the East Indies 
 reference, is about balanced by the tiny probability of 
 before 1619, established by the name of the Company. 
 For charts and lists one would say 1620. 
 
 Sources. 
 
 Perhaps, for the main idea, Lusfs Dominion. See 
 under Date (2). The episode of Romelio's remedial 
 stabbing is from Goulart's Histoires Admirahles, prob- 
 ably in Grimeston's translation (1607) ; a source Web- 
 ster used also for his lycanthropy in The Duchess of 
 Malfi. 
 
 The State of the Play. 
 
 There is no reason to suppose that any part of the 
 play is not by Webster, or that it has been much ab- 
 breviated or revised. The title-page (1623) avows it 
 "the true and perfect copy, from the original." It 
 may be true. But that the original may have borne 
 
APPENDICES 259 
 
 signs of alterations for stage purposes, is suggested 
 by the fact that (pp.126, 127) on three separate occa- 
 sions in III. 3, the 1623 edition has "Surgeon" where 
 it ought to be "Surgeons," for there were two surgeons 
 in the case. It would have lessened the dramatic effect 
 but not hurt the plot to reduce these two to one, and 
 it is just the kind of change that might have been made 
 in order to use fewer actors. Her Majesty's Servants 
 were on the downhill when they acted this play. And 
 if this change was made for acting, others may have 
 been. 
 
Appendix J. — "A Cure for a Cuckold" 
 
 Date, 
 
 A Cure for a Cuckold was published in 1661. 
 
 (1) It is necessary at one point that a sea-fight 
 should have taken place and be narrated. The Enghsh 
 merchant-ships are reported to have been attacked by 
 three Spanish men-of-war, off Margate. From its style 
 this play must date from the end of James', or from 
 Charles', reign. At any period the dramatist would be 
 likely to attribute fighting, in a play of contemporary 
 life, to the actual enemies of England of the time ; and 
 at this period he would be especially unlikely to offend 
 by suggesting enmity with any friend of the rulers of 
 the country. So we may find it probable these lines 
 were written between 1624 and 1630 (inclusive), when 
 England and Spain were at war; not earlier, while 
 Charles' fantastic matrimonial expedition was going on, 
 and not later, when peace had been patched up. The 
 fact that England was more importantly at war with 
 France from 1627, tends a little to narrow it to 1624- 
 1627. This is a moderate proof of the date of these 
 lines, or one of them; a proportionately smaller one, 
 therefore, for the whole play. 
 
 (2) The plot of "Webster's portion" of A Cure for 
 a Cuckold is the same as, or similar to, that of other 
 plays. It is a particular form of the favourite Eliza- 
 
 260 
 
APPENDICES 261 
 
 bethan motif, Mistress — ^Lover — Friend. On this point 
 I have little to add to and not much to subtract from 
 Dr. Stoll's arguments. The bulk of mine are a sum- 
 mation of his. He seems to me to prove his point ; not 
 as conclusively as he believes ; still, to prove it. 
 
 In giving a synopsis of the relevant parts of the plots 
 of these plays I shall, for clearness' sake, call the pro- 
 tagonist — the lover — A, the friend F, and the Lady L. 
 
 (a) In Mars ton's Dutch Courtezan (1604) L (a 
 courtezan) and F are in love first. F chucks her. L, 
 for revenge, encourages A, who has conceived an over- 
 whelming passion for her; and promises herself to him 
 if he will kill F. A promises to do so; on reflection 
 repents, and warns F. They agree on a trick together, 
 feign a quarrel, and pretend to fight a duel. F hides, 
 and is given out as slain in the duel. To punish A for 
 his folly he hides also from him. L, to complete her 
 vengeance, has A arrested for murder. As A finds he 
 cannot produce F to clear himself, he is in a bad way. 
 At the last moment F, present in disguise, reveals him- 
 self. L is led off^ to prison. A is cured of his passion ; 
 and all is for the best. 
 
 (b) In Massinger's The Parliament of Love (1624) 
 A and L have been contracted in marriage ; A has, im- 
 patiently, first proposed, and then forcibly attempted 
 copulation before the marriage-ceremony ; and L is con- 
 sequently possessed by hatred for him. The tale is told 
 in four scenes. (II. 2) A insists on seeing L and off*ers 
 to do anything she likes to obtain her pardon, and her. 
 
262 JOHN WEBSTER 
 
 She accepts the bargain and bids him find out his best 
 friend and kill him. 
 
 (III. S) A soliloquises that he has tried many friends 
 with a proposal and none of them has turned out a 
 true one. Enter F, who is ecstatic over an unhoped 
 meeting with his mistress, which she has appointed for 
 two hours hence. A is melancholy and tries to slip 
 away. F insists on knowing the reason. A says he has 
 to fight a duel shortly, and can't find a second. F in- 
 sists on coming as second, and cutting his mistress, in 
 spite of A's protestations. 
 
 (IV. 2) They arrive at the duel-ground. A makes F 
 swear to fight relentlessly; then reveals the truth, he 
 himself (A) is the ever detestable enemy. He insists on 
 fighting, is beaten, but not killed. 
 
 (V. 1) It is common talk that A has killed F, and 
 that L has had A arrested for trial before "The Parlia- 
 ment of Love." 
 
 At the trial A is found guilty of murder, L of 
 cruelty, and condemned. L repents and forgives A. F,^ 
 supposed (by a trick arranged, presumably, with A) 
 to be dead, rises from his bier. All is put right, and A 
 and L marry. 
 
 (c) In A Cure for a Cuckold, L (Clare) is secretly in 
 love with F (Bonvile), who has been married, on the 
 morning the play begins, to somebody else. The tale 
 is told in five scenes. 
 
 (I. 1) L is sad. A (Lessingham) renews a previous 
 proposal to her. L will accept on one condition. A 
 
APPENDICES 263 
 
 agrees. L tells him it is to find out and kill his best 
 friend. 
 
 (I. 2) A soliloquises. Enter some friends, and de- 
 mand the reason of A's sadness. A says he must fight a 
 duel next morning at Calais, and has no second ; sec- 
 onds to fight. He asks each to be his second. They re- 
 fuse and exeunt. Enter F ; demands to know the reason 
 of A's sadness. A reluctantly explains. F offers to 
 come, and cut his wedding-night. A protests. F in- 
 sists, in spite of the arrival on the scene of his newly- 
 married wife. 
 
 (III. 1) They arrive at the duel-ground. A says 
 he has come to fight an innocent enemy ; i.e. F, he re- 
 veals. And he is so deep in love, he says, he must kill 
 him. F quibbles that as a "friend" he now is dead. 
 They part. 
 
 (IV. 2) A reports to L F's death. L confesses her 
 unhappy love for F and declares herself overjoyed. A 
 turns against her. 
 
 After some complications with the other part of the 
 plot, 
 
 (V. 2) A and L are reconciled, and marry. 
 
 Before we can proceed to the comparison of these 
 plots there is one point in A Cure for a Cuckold to be 
 got clear. That is, Clare's motive in giving Lessing- 
 ham the command. There are various remarks about 
 it in the play. In I. 2, Lessingham, in his soliloquy, 
 rather meekly wonders "what might her hidden purpose 
 be in this?" He can only suggest that she has a psy- 
 chological interest in proving the proposition that there 
 
264 JOHN WEBSTER 
 
 is no such thing as friendship. In II. 4, Bonvile's ab- 
 sence is commented on. Clare, in an aside, says : 
 
 I fear myself most guilty for the absence 
 Of the bridegroom. What our wills will do 
 With over-rash and headlong peevishness 
 To bring our calm discretions to repentance! 
 Lessingham's mistaken, quite out o' the way 
 Of my purpose, too. 
 
 In III. 1, in the dialogue between the friends, Lessing- 
 ham has a new reason to suggest : 
 
 . . . She loathes me, and has put. 
 
 As she imagines, this impossible task. 
 For ever to be quit and free from me. 
 
 In III. 3. When the news comes that Bonvile is at 
 Calais, as Lessingham's "second," Clare guesses the 
 truth, and cries, aside again: 
 
 fool Lessingham 
 Thou hast mistook my injunction utterly. 
 Utterly mistook it ! . . . 
 
 1 fear we both are lost. 
 
 In IV. 2. Lessingham reports to Clare that he has 
 fulfilled her injunctions. 
 
 Clare. Then of all men you are most miserable : 
 
 Nor have you ought furthered your suit in this. 
 Though I enjoined you to 't; for I had thought 
 That I had been the best esteemed friend 
 You had i' the world. 
 
 Less. Ye did not wish, I hope, 
 
 That I should have murdered you. 
 
APPENDICES 265 
 
 Clare. You shall perceive more 
 Of that hereafter. . . . 
 
 She asks who the slain friend is, and hears "Bonvile." 
 At first she is "lost for ever." Then she suddenly 
 changes and professes great pleasure, promises in- 
 stantly to marry Lessingham, because he has rid her of 
 her "dearest friend and fatalest enemy" — she was in 
 love with Bonvile. 
 
 And beholding him 
 Before my face wedded unto another. 
 And all my interest in him forfeited, 
 I fell into despair; and at that instant 
 You urging your suit to me, and I thinking 
 That I had been your only friend i' the world, 
 I heartily did wish you would have killed 
 That friend yourself, to have ended all my sorrow. 
 And had prepared it, that unwittingly 
 You should have done 't by poison. 
 
 Later, Lessingham turns against her, and leaves her. 
 She, in a soliloquy, expresses great remorse : 
 
 I am every way lost, and no means to raise me 
 But blessed repentance . . . 
 . . . Now I suffer. 
 Deservedly. 
 
 Bonvile appears. She rejoices to find him alive. 
 After some conversation — 
 
 Clare (giving Bonvile a letter) 
 
 . . . had you known this which I meant to have 
 
 sent you. 
 An hour 'fore you were married to your wife. 
 
266 JOHN WEBSTER 
 
 The riddle had been construed. 
 Bon. Strange! This expresses 
 That you did love me. 
 Clare. With a violent affection. 
 
 Bon. Violent indeed; for it seems it was your purpose 
 To have ended it in violence on your friend : 
 The unfortunate Lessingham unwittingly 
 Should have been the executioner. 
 Clare. 'Tis true. 
 
 In V. 2 she again expresses contrition to Lessing- 
 ham: 
 
 Clare. It was my cause 
 
 That you were so possessed; and all these troubles 
 Have from my peevish will original; 
 I do repent, though you forgive me not. 
 
 Dr. Stoll's impression is that Clare's motive is mainly 
 meant to be jealousy of Bonvile (F) and a desire for 
 his death, but that occasionally obscurity comes in and 
 that she seems to have meant something else. As the 
 motive in The Dutch Courtezan was also jealous hatred 
 of F, while that in The Parliament of Love was hatred 
 of A, this tells a little against Dr. StolPs idea that The 
 Parliament of Love came between The Dutch Courtezan 
 and A Cure for a Cuckold. He brings the "obscurity 
 of motivation" into service, however, by an ingenious 
 theory of Webster starting with a plot where the motive 
 was jealousy of F, and introducing phrases and ideas 
 {e.g. "Kill for my sake the friend that loves thee dear- 
 est") from the other. Parliament of Love, motivation 
 of offended modesty. 
 
APPENDICES 267 
 
 But this will not do. It is impossible to imagine 
 that Webster had a mind with so extraordinarily feeble 
 a grasp. And an inspection of the relevant passages, 
 quoted above, shows the truth. Lessingham's own 
 conjectures, of course, are astray. He is meant not to 
 know what Clare is at. The only place which favours 
 the view that her motive was a jealous desire for Bon- 
 vile's death is where she confesses it to him, near the 
 end of the play. If this is true, it is absolutely at vari- 
 ance with the rest of the play, which is perfectly con- 
 cordant with itself. We do not know, at the beg-inningr 
 of the play, that Lessingham's best friend is Bonvile. 
 Nor, as far as we can see, does she. She once says, and 
 once practically admits, to Lessingham, that her com- 
 mand really meant that he was to kill her. And — which 
 far outweighs anything said to another person, for that 
 might be a lie — she twice, in an aside, says that Les- 
 singham mistook her words and is doing something she 
 did not intend. It is perfectly plain and indisputable. 
 She was not aiming at Bonvile. Her remorse for her 
 folly was natural, and does not demand the jealousy-of- 
 Bonvile theory. And her statement to Bonvile must be 
 explained away. 
 
 It might be suggested that it was a desperate lie, and 
 that the whole thing is a bad attempt at subtle psychol- 
 ogy. Or much more probably, that it is an instance 
 of the dangers that lurk for collaboration, especially if 
 it is not contemporaneous ; and that one of the two 
 authors, probably Rowley, misunderstood a part of the 
 plot the other was responsible for, and innocently 
 
268 JOHN WEBSTER 
 
 roused confusion. But I think the severer course of 
 emendation can be shown to be absolutely necessary. 
 
 For if you look at the passage (the last one quoted 
 from IV. 2) you will see it is really impossible that 
 "your friend" can refer to Bonvile, as it seems to. It 
 makes nonsense of the whole passage! For in that 
 case all the information he gets from the letter is that 
 she loves him. And how would that have construed "the 
 riddle .?" For the "riddle" included, by this hypothesis, 
 her queer injunction to Lessingham and its hidden in- 
 tention to end in Bonvile's death ; all of which Bonvile 
 would be ignorant of, an hour before his marriage, and 
 which she'd be scarcely likely to reveal to him ! More- 
 over, what does "unwittingly" mean! How do you 
 kill a man "unwittingly," if you challenge him to a duel 
 in order to kill him.? The whole thing is mad. 
 
 Of course, some small change has to he made in the 
 text. Either "on your friend" must be changed to "on 
 yourself" ; or, more probably, "and" should be read for 
 "on," and the whole should be punctuated : 
 
 "To have ended it with violence ; and your friend. 
 The unfortunate Lessingham^ unwittingly," etc. 
 
 and the whole tale is this. She gives him a letter which 
 he was to have opened just before his marriage. He 
 reads it. It tells him, first, that she loved him. He 
 goes on reading, "Violent, indeed; ... for it seems 
 . . ." It seems, from the letter, that she had intended 
 to "end" (the word fits, by this interpretation) her 
 
APPENDICES 209 
 
 violent love with violence on herself. She was going to 
 have had poison given her. And Lessingham was go- 
 ing to have done it, "unwittingly." She has told Les- 
 singham the whole story five minutes before (p. 309) 
 in the same scene (v. the preceding quotation but one). 
 She even used the same word, "unwittingly." Bonvile 
 was to have learnt of her love and of her death at the 
 same moment, and "the riddle had been construed." 
 
 I have spent some time over this point in order to 
 show that Webster (or Webster and Rowley) is per- 
 fectly clear in his motivation in A Cure for a Cuckold^ 
 and that the motive was this. For it removes the only 
 argument in favour of A Cure for a Cuckold preceding 
 The Parliament of Love; and it may counteract the im- 
 pression that might be produced by Dr. Stoll's harping 
 on Webster's inability to make a plot with coherence 
 or even normal sanity. 
 
 To go back to the comparison of Massinger's, Mar- 
 ston's, and Webster's plays ; when they are summarised 
 in that way, it becomes immediately obvious either that 
 there is some special connection between The Parliament 
 of Love and A Cure for a Cuckold, or that they have 
 a common source other than The Dutch Courtezan: 
 There are so many similarities ; the whole dramatisation 
 of the tale and division into scenes, the "dearest friend" 
 command, the search for him under pretext of asking 
 for a second in a duel, the unsuccessful application to 
 other friends, F cutting his mistress, the duel scene, 
 the supposed death of F, and so on. They cannot pos- 
 
270 JOHN WEBSTER 
 
 sibly have arisen from independent study of Marston's 
 play. 
 
 There may have been an intermediate step, a source, 
 perhaps, in the first twenty years of the seventeenth 
 century, and, if so, probably founded on Marston's 
 play. Dr. Stoll does not consider the possibility of this. 
 But we cannot rule it out. It would explain the general 
 similarity, with such differences of motivation, etc., in 
 Webster's and Massinger's plays. This intermediate 
 source must have been either itself a play or a story 
 that fell very easily and necessarily into certain scenes, 
 as an apparently whole, already carved, chicken drops, 
 as soon as you touch it, into neatly severed limbs. 
 More than this one cannot say. There is little proof 
 for or against an intermediate source. One can only 
 admit its possibility. 
 
 But if only these three plays are left us, which was 
 intermediate. The Parliament of Love or A Cure for a 
 Cuckold? The former is nearer to The Dutch Courte- 
 zan in one point, the law-case at the end, in which L 
 accuses A ; the latter in no point. This is some evidence, 
 but not so strong as it seems, for the law-case at the 
 end of The Parliament of Love is required anyhow by 
 the whole plot, independently of this part. Then there 
 are certain differences in treatment that may be signifi- 
 cant. Webster comments on the strangeness of the 
 seconds having to fight in the duel. Massinger accepts 
 it without comment. Dr. Stoll thinks this a proof that 
 Webster was the later. To me it seems more likely that 
 the inventor of the story should have commented on a 
 
APPENDICES 271 
 
 detail like this, and the man who took the story over, 
 accepted it. Again, Webster directly presents A try- 
 ing several friends in vain before he tries F ; Massinger 
 only relates it. Is it more likely that Webster drama- 
 tised what Massinger reported, or that Massinger made 
 indirect what Webster gave directly .^ The former, I 
 think ; so that this piece of evidence favours Massinger 
 being the intermediary. Dr. Stoll suggests several 
 pieces of more general evidence. (1) A Cure for a 
 Cuckold shows the influence of Fletcher and Massinger. 
 This would have happened if Webster had been imi- 
 tating The Parliament of Love. Therefore he was imi- 
 tating it. (2) Webster could not have invented so 
 dramatic a sequence of scenes himself; and Massinger — • 
 and only Massinger — could. (3) Webster's muddling 
 of motivation shows that he was trying to work The 
 Parliament of Love motives into a different plot. (4) 
 The mass of word-play and quibbling in Webster shows 
 he was, later, an embroiderer. (5) Some of the later 
 invented incidents, e.g., the duel-scene, and also the 
 struggle in A's soul, are Massingerish. 
 
 These are not really at all strong. (1) is bad logic. 
 Webster would have shown — and did show — the influ- 
 ence of the time anyhow. (2) These generalisations 
 about Webster's capabilities, founded on such small 
 data, are very dangerous. Possibly Webster could 
 have invented these scenes. Certainly Rowley, his col- 
 laborator, could. Massinger was not the only person. 
 (3) I have disposed of. (4) has some weight: but as 
 
272 JOHN WEBSTER 
 
 Webster was fond of these queer notions and verbal 
 tricks (he still kept something of his heritage from 
 Donne), and Massinger was less fond, it is not very con- 
 vincing. (5) also has a Httle weight, but it is again 
 dangerous to suppose that Webster and Rowley, writ- 
 ing in the manner of Massinger's period, could not have 
 caught something of that very second-rate magic. In 
 any case the struggle in A's soul comes in The Dutch 
 Courtezan, and ex hypothesi Webster could have used 
 it, even if he hadn't the brains to think of it. 
 
 Parts of some of these arguments, it may also be 
 worth remarking, especially of (2) and (5), depend on 
 The Dutch Courtezan, or something equally remote, 
 being the immediate source of whichever of The Parlior 
 ment of Love and A Cure for a Cuckold was the earlier. 
 
 So far there has been a little evidence of the priority 
 of Massinger's play. Dr. Stoll advances one more 
 proof. He shows the evolution of various fragments of 
 the Dutch Courtezan — Parliament of Love story, 
 through forms that must have been familiar to Massin- 
 ger. To begin with, there is The Scornful Lady (1609, 
 or 10) by Beaumont and Fletcher. Massinger, who was 
 a close student of their work, must have known it. In 
 this play the elder Loveless has forced a kiss in public 
 from the Lady. She condemns him to face the Channel, 
 a year in France, and a French mistress. He goes and 
 soon returns in disguise, to report his own death : which 
 scares her, for a minute, into confessing that she did 
 love him. There is really very little of relevance in 
 
APPENDICES 273 
 
 this : far less than Dr. Stoll makes out.* But it has a 
 certain resemblance to The Parliament of Love. 
 
 The next instance is more interesting. The Little 
 French Lawyer (1619 or 20), by Fletcher and Massin- 
 ger, has a variant of the story. In this, A and F are 
 going, as principal and second, to fight a duel. L gives 
 A a sudden command, which will cause him to cut the 
 duel and sacrifice his friend. There is the struggle be- 
 tween love and friendship, in A's breast. Love wins. 
 This is a curious modification of the other theme; but 
 the similarity is not really great. There are minor de- 
 tails of resemblance, which Dr. Stoll brings out clearly,^ 
 though he exaggerates the main points. Most, at least, 
 of this story in The Little French Lawyer, comes in 
 Massinger's portion of the play.^ 
 
 These two steps do not amount to much, but they 
 help a little. We can see that Massinger's mind was 
 
 * Dr. Stoll's great fault is that he is given to pressing evidence, 
 carelessly and unfairly, in his own direction. He is too eager to 
 prove a case. In this instance, a notable one, he says, that the elder 
 Loveless "elicits" from the Lady, "a rueful declaration, like Leo- 
 nora's in the Parliament of Love, that were he alive she would 
 marry him." It is a concoction of untruths. All the Lady says is 
 that if she had been warned when Loveless was setting out, "these 
 two arms had been his sea." As for Leonora she says nothing of 
 the kind. All she says is that, rather than that Cleremond be exe- 
 cuted and she live and die an anchoress in an eight-foot room built 
 on his grave, she'll marry him. Cleremond is not dead, and nobody 
 thinks he is. Perhaps Dr. Stoll was thinking of Bellisant, who 
 is driven by the supposed death of Montrose to confess she loved 
 him. But that belongs to another part of the plot. 
 
 'Stoll, 168-170. 
 
 *i.e., in Act. I. (C. H. E. L. VI, pp. 139, 9). 
 
274 JOHN WEBSTER 
 
 familiar with variants of the story and similar situa- 
 tions. Since a comparison of his variant and Webster's 
 has also made it seem more likely that Webster imitated 
 him, we may conclude that if The Dutch Courtezan, 
 The Parliament of Love, and A Cure for a Cuckold are 
 the only plays in the matter, that was probably the 
 order in which they were written. The Parliament of 
 Love was licensed in November 1624, so 1625 — is, by 
 this department of the evidence, a probable date. 
 
 We can only say then that this play was very likely 
 written between 1625 and 1642; and rather more prob- 
 ably before 1630 than after. 
 
 QUESTIONS OF AUTHOESHIP 
 
 A Cure for a Cuckold was first printed in 1661 by 
 Kirkman, as by Webster and Rowley. This evidence 
 is of very little value. That Webster's hand is to be 
 found faintly in several parts of the play is shown with 
 probabihty but not certainty, by Dr. Stoll.^ His 
 parallel passages seem to be the only proofs of his thai? 
 have any validity. Beyond this we can say nothing; 
 except that the under-plot, the Compass affair, is prob- 
 ably not by Webster, and certainly might be by Rowley. 
 How much share Rowley or anybody else had in the 
 other part of the play, cannot be settled, at least with- 
 out much more minute investigation than this problem 
 has yet received. Mr. Spring-Rice's and Mr. Gosse's 
 
 » Pp. 37-41. 
 
APPENDICES 275 
 
 subtraction of the main plot of the play, and publica- 
 tion of it by itself (as by Webster), satisfies one's ar- 
 tistic feeling, more than one's desire for correct attribu- 
 tion. 
 
7 
 
BIBLIOGRAPHY 
 
 A, EDITIONS 
 
 B, CRITICISM, SOURCES, ETC. 
 
 A. Editions 
 
 The Works of John Webster. Collected by the Rev. Alex- 
 ander Dyce. Four volumes, 1830. 
 Reprinted 1857, one volume. 
 
 The Dramatic Works of John Webster. Edited by Wil- 
 liam Hazlitt. Four volumes. Library of Old Authors. 
 1857. 
 
 The White Devil and Duchess of Malfy. Ed. Sampson. 
 1904. Belles Lettres Series. 
 
 The Mermaid Series: Webster and Tourneur. Edited by 
 J. A. Symonds, with introduction. 1888. Contains 
 The White Devil and The Duchess of Malfi. 
 
 The White Devil; or, The Tragedy of Paulo Giordano 
 Ursini, Duke of Brachiano, with The Life and Death 
 of Vittoria Corombona, the famous Venetian Curtizan. 
 Acted by the Qwenes Majesties Servants. I6l2, 
 quarto. 
 
 Reprinted 1631, quarto. 
 Reprinted 1665, quarto. 
 Reprinted 1672, quarto. 
 
 Injured Love; or. The Cruel Husband. By N[ahum] Tate. 
 A Version of The White Devil. 1707. 
 
 The White Devil. A Select Collection of Old Plays, vol. 
 iii., Dodsley. 1744. 
 Reprinted 1780. Notes by Isaac Reed. Dodsley. 
 
 The White Devil. The Ancient British Drama, vol. iii. 
 Edited by Sir Walter Scott. 1810. 
 277 
 
278 JOHN WEBSTER 
 
 The White Devil. A Select Collection of Old Plays. Notes 
 by Reed, Gilchrist, and the Editor (Collier 1825). 
 
 Vittoria Corromhona, ou le Diable Blanc. Contemporains 
 de Shakespeare. J. Webster et J. Ford. Traduits 
 par Ernest Lafond. Paris, 1865. 
 
 Vittoria Coromhona. Alt-englisclies Theater. 
 
 Herausgegeben von Robert Prolss. Two volumes. 
 Leipzig. 
 
 The Tragedy of the Dutchesse of Malfy. Quarto, 1623. 
 Reprinted 1640, quarto. 
 Reprinted 1678, quarto. 
 
 The Duchesse of Malfy. 1660 (circa) , quarto. Probably 
 reprint of first Quarto (v. Sampson, Webster's White 
 Devil, etc., p. 404). 
 
 The Unfortunate Dutchesse of Malfy; or The Unnatural 
 Brothers. Written by Mr. Webster. 1708, quarto 
 (printed from 1678 edition for stage purposes). 
 
 The Fatal Secret. A Tragedy. As acted at the Theatre 
 Royal. By Lewis Theobald. A version of The Duch- 
 ess of Malfi, 1735. (Preface avows Webster's author- 
 ship.) 
 
 The Duchesse of Malfy. The Ancient British Drama. 
 1810. (Printed from 1670 Ed.) 
 
 The Duchess of Malfi. Tallis's Acting Drama, part i., 
 1851. (Altered and expurgated.) 
 
 Die Herzogin von Amalfi. Shakespeare's Zeitgenossen imd 
 ihre Werke, 1858-1860. F. von Bodenstedt. ("The 
 Duchess" is translated, the other plays summarised.) 
 
 La Duchesse d* Amalfi. 1865. (Translated by Lafond.) 
 
 The Duchess of Malfi. The Works of the British Drama- 
 tists. John Keltic. 1870. 
 
 The Duchess of Malfi. Dick's standard Plays. 1883. 
 
 The Duchess of Malfi. The Best Elizabethan Plays. Ed- 
 ited by W. R. Thayer. 1892. (Expurgated.) 
 
 The Duchess of Malfi. Temple Dramatists. Edited C. 
 Vaughan. 1896. 
 
 The Duchess of Malfi. Ed. W. A. Neilson. 
 
 The Malcontent, By John Marston. l604. 
 
BIBLIOGRAPHY 279 
 
 The Malcontent. Augmented by Marston. With the addi- 
 tions played by the King's Majesty's Servants. Writ- 
 ten by John Webster. l604. 
 
 The Famous History of Sir Thomas Wyat. With the Coro- 
 nation of Queen Mary and the Coming in of King 
 Philip. As it was plaid by the Queens Majesties' 
 Servants. Written by Thomas Dickers and John Web- 
 ster. 1607. Reprinted in Two Old Plays. Ed. W. J. 
 Blew. 1876. 
 
 North-Ward Hoe. By Thomas Decker and John Webster. 
 1607. 
 
 West-Ward Hoe. Sundry Times acted by the Children of 
 Paules. Written by Tho. Dekker and John Webster. 
 1607. 
 
 The DeviVs Law-Case; or. When Women goe to Larv, the 
 Devil is full of businesse. A new Trage-comedy by 
 John Webster. 1623. 
 
 Appius and Virginia. A Tragedy by John Webster. 1654. 
 
 A Cure for a Cuckold. A Pleasant Comedy. Written by 
 John Webster and William Rowley. I66I. 
 
 The Thracian Wonder. A Comical History by John Web- 
 ster and William Rowley. I66I. 
 
 Love's Graduate. A Comedy by John Webster. 1885. 
 Edited by Stephen Spring-Rice. With an introduc- 
 tion by Edmund Gosse. 
 
 [Merely A Cure for a Cuckold with the sub-plot, the 
 Compass story left out.] 
 
 Monuments of Honor. Derived from remarkable antiquity 
 and celebrated ... at the sole munificent charge . . . 
 of the Worshipfull Fraternity of the Eminent Mer- 
 chant Taylors. Invented and written by John Web- 
 ster, Merchant Taylor. 1624. 
 
 A Monumental Columne erected to the living Memory of the 
 ever-glorious Henry, late Prince of Wales. By John 
 Webster. I6l3. 
 
280 JOHN WEBSTER 
 
 B. Criticism, Sources, etc. 
 
 Retrospective Review. 1823. John Webster [anonymous 
 
 article]. 
 A True Relation of the deserved Death of that base and 
 
 Insolent Tyrant, the most unworthy Marshall of 
 
 France. . . A True Recital of those things that have 
 
 been done in the Court of France since the Death of 
 
 Marshall D'Ancre. I6l7. 
 J. Q. Adams, Junior. Greene's Menaphon and The Thra- 
 
 cian Wonder. Modern Philology, iii. I906. 
 W. Archer. Webster, Lamb, and Swinburne. New Review, 
 
 1893. [Decrying Webster and the "Lamb Tradition" 
 
 about the Elizabethans.] 
 M. Bandello. Novelle. 1554-73. 
 F. de Boaistuau and P. Belleforest. Histoires Tragiques. 
 
 1580-2. 
 J. le G, Brereton, IQOQ. Elizabethan Drama. 
 
 [Contains revised version of the review of Prof. 
 
 Sampson's The White Devil and The Duchess of 
 
 Malfi, which appeared in Hermes, November 1905.] 
 J. le Gay Brereton. The Relation of the Thracian Wonder 
 
 to Greene's Menaphon. Mod. Language Review, Oc- 
 tober 1906. 
 
 Englische Studien. xxxvii., 1907. 
 Crawford, C. Collectanea. Two Series. 1906-7. 
 
 Series I., pp. 20-46. Webster and Sidney. 
 
 Series IL, pp. 1-63, Webster, Marston, Montaigne, 
 
 Donne, etc. 
 
 Reprinted from Notes and Queries. 
 Gnoli, D. Vittoria Accoramboni. 1870. 
 Gosse, Edmund. Seventeenth Century Studies, 1883. John 
 
 Webster. 
 Greg, W. W. Webster's White Devil. Modern Language 
 
 Quarterly, Dec. 19OO. 
 Grimeston. Translation of Goulart's Histoires admirables 
 
 et memorables de nostre temps. I607. 
 
BIBLIOGRAPHY 281 
 
 Hatcher, O. L. Sources and Authorship of the Thracicun 
 Wonder. Modern Language Notes, vol. xxiii., 1908. 
 ^ Hazlitt, William. Lectures on the Dramatic Literature of 
 the Age of Elizabeth. 1821. 
 
 Hunt, Mary Leland. Thomas Dehher. Columbia Uni- 
 versity, 1911. [v. Appendix D.] 
 
 Kiesow, K. Die verschiedenen Bearbeitungen der Novelle 
 von der Herzogin von Amalfl des Bandello in den Lit- 
 eraturen des xvi. und xvii. Jahrhunderts. 1895. 
 
 Lamb, Charles. Specimens of the English Dramatic Poets. 
 1808. 2nd Ed., 1813. 
 
 Lauschke, Johannes. John Webster's Tragodie Appius and 
 Virginia. 1899- 
 
 Lee, Sidney. John Webster. Dictionary of National 
 Biography, 1899- 
 I^Lowell, J. R. The Old Dramatists. Webster. 1892. 
 
 MacCracken, L. A Page of Forgotten History. 1911- 
 {i.e. The Story of Vittoria.) 
 
 Meiners, Martin. Metrische Untersuchungen iiber den 
 Dramatiker, John Webster. 1893. 
 
 Michaelis, S. The Admirable History of the Passion and 
 Conversion of a Penitent Woman, seduced by a Magi- 
 cian. 1613. 
 
 Murray, J. Tucker. English Dramatic Companies, 1585- 
 1642. 2 vols., 1910. 
 
 Painter, W. Palace of Pleasure. 1566. Two volumes. 
 
 Pierce, F. E. The Collaboration of Webster and Dehher. 
 1909. Reviewed by Dr. P. Aronstein in Beiblatt sur 
 Anglia, 1910, p. 79- 
 
 Rich, Barnabe. New Description of Ireland. I6IO. 
 
 Scheffler, W. Thomas Dehher als Dramatiher. 1910. 
 
 Sidney, Sir Philip. Arcadia. 1590. 
 
 Stoll, E. E. John Webster. The Periods of his Worh. 
 1905. Reviewed by W. W. Greg in Modern Language 
 Review, Oct. I906. 
 
 Swinburne, A. C. The Age of Shahespeare. 1908. John 
 Webster, 
 
282 JOHN WEBSTER 
 
 Swinburne, A, C. Prologue to the Duchess of Malfy. 
 
 Nineteenth Century, 1899- 
 
 Studies in Prose and Poetry. John Webster, 1894. 
 
 Symonds, J. A. Italian Byways. 1883. Vittoria Acco- 
 
 ramboni. 
 
 Reprinted in his Studies in Italy and Greece. Second 
 
 Series. 
 The Repentance of Nathaniel Tindall that hilled his mother. 
 
 July 2, 1624. 
 
 A most bloody and unmatchahle murder committed in 
 
 Whitechapel by Nathaniel Tindall upon his own 
 
 mother, written by John Morgan. 
 
 (I cannot find either of these in the British Museum. 
 
 They may not be extant.) 
 Vaughan, C. E. Tourneur and Webster. Article in Cam- 
 bridge History of English Literature, vol. vi. 
 
 Reviewed, Aronstein in Beiblatt zur Anglia, April 
 
 1911. 
 Vopel, C. John Webster. 1888. 
 Wurzbach, Wolfgang von. Jahrbuch der deutschen 
 
 Shakespeare-Gesellschaft. 1898. John Webster. 
 
RETURN CIRCULATION DEPARTMENT 
 TO^^- 202 Main Library 
 
 LOAN PERIOD 1 
 HOME USE 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 
 
 1 -month loans may be renewed by calling 642-3405 
 
 6-month loans may be recharged iDy bringing books to Circulation Desk 
 
 Renewals and recharges may be mode 4 days prior to due date 
 
 DUE AS STAMPED BELOW 
 
 MAY 2 1980 
 
 .Rm JUN 3 iiBO 
 
 JUN 1 4 1988 
 
 RECEIVED BY 
 
 CIRCULATION DEPT 
 
 UNIVERSITY OF CALIFORNIA, BERKELEY 
 FORM NO. DD6, 60m, 1 1 /78 BERKELEY, CA 94720 
 
 (Qll73si0)476-A-32 
 
 .GeneraJ Library 
 
 UmvemtypfCaliSrnia 
 Berkeley 
 
r 
 
 GENERAL LIBRARY • U.C. BERKELEY 
 
 I 
 i 
 
 I'J: 
 
 Boooibbvai 
 
 'fr-r: 
 
 i