[In the following essay, Pierce attempts to reconcile contradictory interpretations of The Tempest by reexamining the meaning of the play.]

Some years ago I wrote an article¹ on Shakespeare's The Tempest in which I gave a reading that was quite sympathetic to Prospero and Miranda. I tried in that article to express an important part of my understanding of the play at that time, and I would still stand behind what I then said. On the other hand, I find much that is appealing and persuasive in a series of New Historical and anticolonialist readings of the play,² which tend to be not at all sympathetic to Prospero and if anything to turn Caliban into a sort of hero or at least victim. How am I to explain this weak-minded doubleness in my understanding of The Tempest? Is it that I cannot make up my mind about what Shakespeare's play means, or can two contradictory readings be part of the way I understand it? My difficulty with The Tempest is suggestive of a larger problem in assimilating varied interpretations of a complex literary or dramatic text, one that I suspect we all encounter.

The task of adjudicating among different and even contradictory commentaries is dependent on what counts as understanding The Tempest. A stern New Critic would argue that all the writing about the New World and English colonialism is irrelevant to understanding the play, however accurate it may be as history. On the other hand, the New Historicist would condemn a formalist commentary on images and characters as naive and shallow, as incapable of addressing the social and political implications of the text. The two point in very different directions where to look in seeking the meaning of The Tempest. What might it mean to say that one understands a play? And what is the function of critical commentary in providing that understanding? These seem like very basic questions for our enterprise of teaching and writing about literature, including drama, and no doubt we all have developed some personal answers to them, but they are hard to deal with in any definitive way. I would like to suggest that the difficulty comes partly from a misleading picture of what constitutes the meaning of a play, the kind of misleading picture that according to Wittgenstein bewitches much of our thinking.³ This article will not solve all the problems of how to understand a play, but I do hope to sort out some of the issues and confusions in a way that will clarify how to make critical commentary useful and how to judge the usefulness of the criticism we read.

I plan to take as my test case Shakespeare's The Tempest, partly because the flood of recent commentary on it has raised much controversy and for many of us has radically altered our understanding of the play. How can one play look so different from different perspectives, and how can I make sense of my seeing it in two such seemingly incompatible ways as the traditional and colonialist readings? Should I reject the one or the other view as mistaken or perverse? Can I reconcile them in some larger framework? Or must I simply live with the incongruity?

Let me suggest the nature of the problem by quoting a series of statements about The Tempest, all of which presumably make some kind of claim to truth and to usefulness:

In The Tempest whatever evil remains is impotent, and goodness returns to action. Here, as in all the last plays, there is a re-birth, a return to life, a heightened, almost symbolic, awareness of the beauty of normal humanity after it has been purged of evil—a blessed reality under the evil appearance.⁴

In Prospero's metastance striving for power, secure identity, and certain belief is transcended through a choice by the whole self to live with faith in a world it knows man can never fully control or predict.⁵

To come to the island is to start life over again—both his own and Miranda's—with himself as sole parent, but also with himself as favorite child. He has been banished by his wicked, usurping, possibly illegitimate younger brother Antonio. This too has the shape of a Freudian fantasy: the younger child is the usurper in the family, and the kingdom he usurps is the mother. On the island, Prospero undoes the usurpation, recreating kingdom and family with himself in sole command.⁶

Caliban's retort might be taken as self-indictment: even with the gift of language, his nature is so debased that he can only learn to curse. But the lines refuse to mean this; what we experience instead is a sense of their devastating justness.⁷

Prospero needs Miranda as sexual bait, and then needs to protect her from the threat which is inescapable given his hierarchical world—slavery being the ultimate extension of the concept of hierarchy.⁸

Leontes and the others recognize the nobility of Miranda in The Winter's Tale though they do not know who she is.⁹

The descriptions that these quotations represent surely incorporate or imply some ideas about the play that are inconsistent with one another. Prospero is an admirable figure; he is a self-deceived tyrant and manipulator. The play portrays a victory over evil; the play is a grim anatomy of human ruthlessness and cruelty. Shakespeare is a wise and Olympian observer of the human condition; Shakespeare is the conscious or unconscious servant of a hegemonic state apparatus. In addition the writers illustrate heterogeneous ways of approaching the play, different critical methods which themselves imply judgments about the tools involved: is psychoanalytical thought a better way of understanding human behavior than a commonsense, traditional psychology of purposes and intention? Should we take for granted the author's wisdom or look for signs of authorial implication in bigotry and oppression? Is it important to be rational and accurate, even to be clear what play one is talking about, or is the free play of imagination everything?

How we deal with these contradictions is dependent on how we define the goal of finding the meaning of The Tempest.¹⁰ Is that meaning some set of words (other than those of the play itself), presumed to represent what was in Shakespeare's head or what would be in the head of an ideal reader or viewer: the play's comment on life or something like that? Or perhaps the meaning is some idea not to be expressed in words, something deeply ineffable. At any rate, what The Tempest means in this view is a group of ideas, whether expressible in language or not. That picture of meaning as a set of words or thoughts is, I would contend, implied in much of our thinking about drama, even though we might find the picture a bit naive when described so baldly. And such a picture of dramatic meaning seems to me misleading: not wholly wrong, but significantly distorted as a mapping of how we understand a play.

The very term “the meaning of a play” pushes us toward an image of that meaning as a text or mental diagram of some sort, an entity which we think of as both the goal of our attention to the play and our means of understanding it. We read the play in order to get the meaning, but we also use our preliminary sense of the meaning as a tool to find our way through the play. Again, such a description of a play's meaning has considerable truth as an indication of how we behave as readers and viewers, but it is only a map of what goes on when we seek and find meaning. Maps by their nature indicate some parts of what they represent very well while omitting or distorting others. It is no shame to a road map that it fails to indicate the contours of hill and valley or that it represents towns as circles despite their actual variety of shapes, but the usefulness of such a map is limited by its design to represent some things and not others. It can tell us which road to take but not the altitude of some section of that road. In the map I have sketched above of how a play means, what is ignored or misrepresented?

The answer is that this map implies the existence of a thing, the meaning of the play. In our use of language we have to learn over and over again that nouns do not necessarily represent things and thus that there may be no entity, not even an abstract entity in the mind, to which the term “dramatic meaning” refers. Let me suggest an alternative picture that I think maps certain qualities of dramatic meaning better than the traditional one as I have described it. Understanding the meaning of a play is not so much like grasping a concept as like having an ability, for example, the ability to polka. I learned in high school to polka, and I believe that I can still more or less do it, though I have not tried for years. Is that ability to polka some entity in my mind, some consciousness (or some unconscious thought)? I do not think so. Presumably I have the ability even at moments when I am not thinking of polka-ing: I am a person who can polka even though I am now writing at the computer keyboard. And my ability is not to be seen as a pattern in my head approximating some ideal pattern, the perfect polka. That idea of a mental pattern would describe well enough my grasp of the fifty states of the United States. I seem to have some sort of trace in my memory of the fifty states, a mental set which matches the list of the states in a history book. When I do remember the fifty states, my list can be lined up with the list in the book state for state. On the other hand, I may get better (or worse) at polka-ing, but my improvement does not manifest itself in getting closer to that hypothetical ideal, the idea of the polka, so that I would look more like you as both of us approach the ideal. In short, there is no entity, polka-ing, of which I have a more or less accurate copy in my head for my body to match up to when I know how to polka.

How then do I—and how do you—decide that I can polka with some skill? The answer is that there are criteria, an indefinitely large set of rules of thumb, by which we make such evaluations.¹¹ In the first instance I need no criteria: I just know with some degree of confidence that I can polka as I know that I can throw a baseball or speak English. You presumably believe me when I claim to have that ability. Thus one main criterion for you is my avowal of being able to polka, what I said in the previous paragraph. But of course I could be mistaken in my belief, or I could be trying to deceive you, perhaps even trying to deceive myself. And so there are other criteria available in case you and I want to be more careful in our assessment of my skill. Most obviously, I could try to polka, and you could watch me: am I tripping over my feet or my partner's? Am I keeping time? How gracefully do I move? Notice that these criteria vary in their objectivity. Keeping time is more less an exact matter: does my footfall coincide with the musical beat? But grace is a matter of judgment: we learn by experience to evaluate it, and some people have a better eye for it than others.

We could pretend that the concept of knowing how to polka is meaningless because we cannot define it in clear terms, cannot find the entity that constitutes it. But the would be only a philosophical quibble. Our society gets along quite well deciding whether or not people can polka. No doubt there are imprecisions at the margin in such judgments: indeed I am not all that confident that I can still polka. Also I am sure there are people, even among those familiar with the dance, who would make peculiar evaluations of people's skills at polka-ing. But imprecision does not invalidate a concept; the appropriate test is its workability, so that we can communicate with it in our social interactions. Indeed imprecision around the edges is often valuable, as when we find it convenient to use a vague adjective like “close” rather than using only a language of exact measurements. I can say, “You are too close to that cliff edge,” when I lack the information to say, “You are.43 meters from that cliff edge,” and you will probably find my information useful despite the imprecision.

I want to claim that understanding the meaning of The Tempest is more like having an ability—like being able to polka—than like knowing the fifty states or even the structure of the United States government. What are the implications of that statement? Let me begin with some negatives. First, the meaning of The Tempest is not just a series of true statements about it. Thus there is no ideal article, or even book, toward which all the actual critical writing in the world is or should be evolving; there is no possible text that is the meaning of the play. Of course an article should try to make true statements and avoid false ones, and the true statements that it makes may well be useful in helping the reader to understand the meaning, but they will not be sufficient.¹² My making all sorts of true statements about The Tempest is no guarantee that I understand its meaning, and indeed no set of true statements that I made could provide such a guarantee.

Second, the meaning is not to be seen as some idea or image in the author's mind, what we often equate with the author's intention. Even if we assume that some Polish genius invented the polka, my ability to polka is not my capacity to mirror a picture in his or her mind—the polka that the inventor created mentally before telling the rest of us about it. The same principle holds for a play: the meaning is not an entity in the author's mind. Information about the author's intention may well be valuable. Only the most dogmatic of New Critics would refuse to take seriously Milton's claim in Paradise Lost that he aims to “assert Eternal Providence, / And justify the ways of God to men.” And Prospero tells us that his aim—as character, actor, and perhaps also spokesman for Shakespeare—“was to please.”¹³ Both of these thoughts were presumably in the authors' minds at some point, and surely they are relevant to understanding the two works, but they do not constitute the meaning in either case.

One might take this description of meaning to suggest that the meaning of The Tempest is multiple—for example, that the play is both a defense of patriarchal hierarchy and a subversion of its pretensions. But that is like saying that there are several ideal polkas rather than just one. Why is that an inadequate picture of the meaning of The Tempest? One main reason is that there are aspects of understanding the play that do not lend themselves to expression as ideas. Would we say that someone who had no sense of the figures of Ariel and Caliban, who did not respond to the poetry of the songs, understood the meaning of the play? Indeed the picture of the mind as a container holding a set of ideas is deeply unsatisfactory, as Wittgenstein demonstrates at length in Philosophical Investigations. Thus the third negative implication is that there is no thing, no entity or group of ideas, whether in the author's mind or the reader's or viewer's, that constitutes the meaning of The Tempest. Ideas are a part but not the whole of the meaning.

Fourth, there is no goal, no endpoint, to the process of coming to understand the meaning of the play. In playing games, there are two characteristic patterns.¹⁴ In one the player tries to attain some goal: getting to the top of the mountain, scoring more points than an opponent. In the other the player just tries to carry out some task better and better: skating, finding words in a group of letters. I contend that seeking the meaning of The Tempest is like the second sort of game: since there is no entity, the meaning, there is no one goal line. I can always keep getting better at understanding the play, and no leap of insight is the final truth. But better and worse are perfectly meaningful concepts here. I may well understand The Tempest better after reading the articles from which I have quoted, even though that understanding does not bring me closer to some one ideal body of ideas, just as I get better at polka-ing without approximating some ideal polka.

But if there is no goal line for understanding the play, how do I know how well I am doing, and how do you judge me? That is where the concept of criteria is helpful. Just as we judge my ability to polka by criteria, so we judge my understanding of The Tempest. Perhaps the most basic of these criteria for me individually is the click of enlightenment,¹⁵ when I get that sense of having seen of having seen the point. It may be sudden or gradual, but at its most striking it is like a light being turned on in a dark room. Now the elements of the play fall into place, make sense. It is above all that experience that leads me to make an avowal of understanding, which is the primary basis for your deciding that I understand the play. If you were to ask me whether I understand The Tempest, I would with a few modest qualifications answer yes (as I would not for Ulysses); and if you considered me trustworthy, you would probably believe me.

But how can we be sure that I am not mistaken? Or what if I am not so sure about making the avowal? The light may seem a bit dim, so that I am not sure what and how much I see. I can test the validity of my own sense of understanding by trying it out: can I see connections among different parts of the play? Do elements that have confused me in the past now fall into place? Can I persuade other people of my insights? Can I answer their questions about the play? And other people use similar criteria to judge my claim to understanding. Beyond that they can listen to my reading-aloud of parts of the play: does my voice make sense of the lines? Does it sound plausible as Prospero and Miranda and Caliban? No one of these criteria is definitive, and no one yields indisputable results. And yet their cumulative weight is powerful because it is what we mean by understanding the play, the way we have learned to apply the expression.¹⁶

But surely, you may well be saying, this is perverse. Either there is a meaning of The Tempest, or there is not. There may not be any single set of words that constitutes that meaning, but sets of words may approximate it more and more exactly. If that were not the case, then I would be forced back on total relativism: The Tempest means whatever I think it does at any moment, which is much the same as saying that it has no meaning, that meaning is purely an arbitrary creation of the individual (or group) consciousness.¹⁷

I would contend that that choice between one specific meaning and total relativism is a false dilemma. After all, it is not true that either there is one ideal polka or else anything counts as a polka. We can judge whether given behavior is polka-ing by criteria, and that fact is sufficient to make the concept of the polka meaningful. But it may well be more accurate to say that The Tempest has meaning than that it has a meaning or meanings. The latter expressions create a misleading picture, in that they suggest an entity or entities in Shakespeare's head or yours or mine that constitutes a standard. But no one has explained what sort of thing that entity would be, and anyway it is unnecessary in explaining how we apply the concept of meaning.

Why does this distinction between meaning as entity and meaning as area for the deployment of skill make any difference? In part it helps to clarify what we look for when we seek meaning. First, it makes clear that there is no one unified thing to be sought. Parts of what we count as meaning may be disconnected from one another in our perception. What does this suggest about the critical concept of unity? Surely unity is a valuable concept, one that leads us to new insights as we try to apply it to a text. Yet we really know that not everything about a text is unified. There are attributes that are of no importance—the number of occurrences of the letter “a” is usually one such—and other attributes may be significant in independent ways—a speech may be a funny set piece not closely connected to the rest of the play. Indeed there is no a priori reason to deny that a text may combine incompatible assertions: The Tempest may well both support and undermine patriarchal hierarchy. The fact that not everything fits together in the play is not a disastrous flaw, nor is it the indication of some deeper meaning to be excavated or just postulated as really there despite our incapacity to see it.

Second, meaning may and indeed usually does include elements other than assertions. Comprehension of a character is the most obvious example of that: surely grasping the sort of human being that is Prospero counts as part of understanding the meaning of the play. We might be able to express part of our grasp as a series of statements about him, but that would be an awkward approach to what we do largely by intuitive means: we all know actors who can play a character brilliantly but are blankly inarticulate in trying to describe what they are playing, just as there are people who read other people acutely without being able to express their understanding in psychological language. Will we grant the actor and the shrewd people no understanding of the meaning of the character or person?

On the other hand, there are critical utterances about Prospero which sound to me like possible statements about some play called The Tempest, but which I cannot match up with any Prospero that I, at least, can imagine in Shakespeare's play. Thus I cannot really see Shakespeare's Prospero as thwarting Antonio out of jealousy of the younger brother who has usurped their parents' love. I cannot fit Orgel's description against what I see Prospero say and do in the play. That is for me an insuperable objection to accepting such a critical utterance as useful in understanding the play, though at some future date someone might be able to talk me into seeing such a Prospero, or an actor might be able to show it to me. I believe that Orgel describes a real phenomenon in the world—older brothers can feel such jealousy and act on it; but the idea does not help my grasp of Shakespeare's Prospero. My intuitive sense of what assertions are congruous with the character is surely part of how I grasp his meaning. I would sum up this second point in an apparent redundancy: the meaning of a play is what I try to understand when I try to understand the play. Whether or not an element of that understanding is exactly an idea, something that can be thought of as an assertion and expressed in language, it is still part of the meaning. But an idea that does not help in understanding the play is not part of the meaning no matter how brilliant as an observation of the world or seventeenth-century England or whatever.

Third, meaning is not independent of the reader or viewer. Since there is no entity that constitutes the meaning, there is no one objective thing toward which both you and I are moving. In a sense you and I have to climb our own mountains, though we can benefit greatly from each other's insights and guidance. But what helps you to understand the play may not help me, even though I have no objections to it on grounds of truth or relevance. If I simply cannot see The Tempest from Caliban's perspective, then I cannot incorporate that into my understanding of the play, no matter how helpful it is to you. No doubt I would understand the play better and more fully if I could do so, but that step is further up my mountain than I have yet climbed.

How does this set of observations about dramatic meaning affect my response to the critical quotations at the beginning of this article? Spencer in the first excerpt identifies a traditional theme in the play, rebirth, and he makes a traditional dichotomy between good, associated with Prospero and his project, and evil, associated with Caliban and the plotters. One could offer a great deal of evidence for finding the theme of rebirth throughout the last plays, as many critics have,¹⁸ and thinking of it as a theme¹⁹ helps to make sense of the imaginative power in such elements of the play as the shipwreck that turns out to be imaginary, Ariel's song “Full fathom five,” and the revival of Alonso, Gonzalo, and the others from the magic spell cast on them. I suspect that almost any reader or viewer who has not yet thought of the theme of rebirth feels a click of recognition, of elements of the play falling into place, when encountering the idea in Spencer or Tillyard or someone else.

More controversial is Spencer's implied judgment of Prospero as associated with good, even embodying it. Leininger, for example, sees him as an embodiment of patriarchal and colonialist manipulation. In some respects the two critics are agreeing about what they see but judging it differently: Spencer finds acceptable and even admirable the very exercise of manipulation and control by a patriarchal figure that Leininger finds offensive. But her sentence includes a somewhat ambiguous claim, all forms of which I think Spencer would dispute. First, it may mean that in the incident referred to at 1.2.353 ff. Prospero has consciously used Miranda's beauty to impel Caliban to attempted rape so that Prospero will be justified in enslaving Caliban. Second, it may attribute such a motive to him unconsciously. Third, it may contend that the social context of patriarchalism and colonialism incorporates this pattern of forces independent of Prospero's intention, conscious or unconscious. Ideology is shaping Prospero's and Miranda's attitudes, but the causal link occurs in the development of the social process, not in Prospero's mind. He idealizes her virtue because he has picked up the values of a social group that needs to justify enslaving other males and so to perceive them as rapists.

Should we take Prospero's—and Miranda's—explanations of their original kindly behavior toward Caliban at face value, as Spencer does, or should we seek for the real reason in Leininger's mode? The first and second versions of her interpretation seem to me dramatically implausible. They purport to know things about an undramatized incident for which the play simply does not offer evidence, and so they invent a past for Prospero just as Orgel does in his quotation. One could imagine a play dramatizing Prospero's behavior toward Caliban in those terms, but that imagined play is remote from the actualities of The Tempest. One way of making this point is to ask how the actor playing Prospero could suggest having had this devious motive for letting Caliban live with them, whether by the way he delivers the lines at 1.2.346 ff. or by behavior elsewhere. The psychology of Prospero that these versions offer is out of keeping with the impression I derive from the rest of the play. Whatever one may think of his educative methods as we see them, he is not using them to undermine Caliban's character.

If, as I suspect, Leininger is really making the third assertion, the issue is more complex. First the claim rides on the justifiability of seeing Prospero as a colonialist, which has to many recent critics seemed almost inevitable.²⁰ It is of course an allegorical interpretation, since literally the island is in the Mediterranean and Caliban is not an Indian. And the colonialist reading is not the only allegorical interpretation that has seemed compelling to critics: Prospero has also been seen frequently as Shakespeare himself, as the artist or dramatist in general, and as God. Mark Van Doren comments, “Any set of symbols, moved close to this play, lights up as in an electric field. Its meaning, in other words, is precisely as rich as the human mind, and it says that the world is what it is.”²¹ The specific allegorical reading involving colonialism lights up in just that way for our generation. I myself doubt that Prospero is allegorical of the colonialist in the way that Spenser's Faerie Queene stands for Elizabeth: Spenser consciously intended his equation, as the letter to Ralegh makes clear; and the poem is full of signs that point unambiguously in that direction. One might call the reading of Gloriana as Elizabeth a compelled allegory, which the interpretation of Prospero as planter, or as Shakespearean dramatist, is not. But I suspect that most modern readers and viewers feel something like the sentiment voiced by Greenblatt in his quotation, at least at the moment of the play he refers to (1.2.365 ff.), and that is because Caliban is so recognizably in the position of the colonized subject. For us who respond that way, understanding the play involves acknowledging that feeling and making sense of it. Not to feel the force of analogy between Prospero's island and the colonial enterprise is not to be very much at home in either the seventeenth century or the end of the twentieth. I suspect that most who resist the analogy are simply unwilling to move out of the older reading, whether through ideological resistance or through sheer inertia.

A second premise for justifying this version of Leininger's argument is that it represents a plausible account of a recurrent political process: fathers and colonialists behave as she represents Prospero behaving. Leininger argues that the process is represented in the historical reality of the Princess Elizabeth, for whose nuptials the play was performed in 1613: here was an actual woman used as political bait. But is it specifically the case that enterprises of subjugation, especially colonialist ventures, frequently justified themselves by tempting the people to be subjugated with women of the ruling class and then punishing them with enslavement for responding to the sexual temptation? I myself have doubts about the idea since I think it assumes more psychological need for self-justification than most subjugators have felt. Besides, I wonder if it does not extrapolate too far from the psychosexual attitudes of American slave holders and the postbellum defenders of Jim Crow. But those who find the historical theory more plausible than I do may well find explanatory power for Prospero's behavior in an ideology that glorifies virginity and demonizes the subjugated as threats to that virginity. He need not have any plan to entrap Caliban through Miranda; he acts in a way that produces that result because of attitudes shaped by an ideology that exists to justify hierarchy and subjugation.

Spencer and Leininger imply opposite evaluations of Prospero, and indeed nearly all interpretations of the play tend in one or the other of these two directions, though with various qualifications. Is arriving at a judgment of him part of understanding the play, or is it simply a case of de gustibus, one of those value judgments that our students scorn? I would suggest that there is no good reason to doubt the cognitive status of such judgments and indeed that one of the main purposes for reading critical interpretations is to guide us in our own arrival at a considered judgment of Prospero. In a sense we can try out the different perspectives, see how well they work in our process of thinking about him.

In some versions the different evaluations of Prospero may be logically or psychologically incompatible. I can think of him as a kindly old dramatist giving up his art after one last performance by the creatures of his imagination, or I can think of him as a manipulative ruler and colonizer who inflicts pain and pleasure to suit his own plans, but I cannot see both Prosperos at the same time; the two images replace each other like Jastrow's duck and rabbit.²² One element in this dichotomy is that I have to think of characters like Ariel and Caliban either as representing human beings or not—perhaps rather as elements of Prospero's own psyche or as aspects of dramatic creation. Prospero looks very different depending on whether I think of his slaves as human beings. Nevertheless, I may see Prospero in those two aspects as two different potentialities of the play, which can, for example, be brought out in different productions. In a sense I am understanding two plays instead of one, but why not accept that possibility? On the other hand, not all such contradictions demand two different Prosperos. To understand a character is not to find some single explanatory formula, the idea of the character, any more than that is true of the play as a whole or of a human being outside drama. Again, the formulae for Prospero—colonialist oppressor, benevolent father, and so forth—are not the goals of understanding but the tools.

Some critics may feel that it is my duty to accept only the colonialist version of Prospero and the play, to see entirely from Caliban's perspective, as it were. That demand may be a kind of dour secular Puritanism: any time when I am not thinking about political oppression is time wasted on unpolitical frivolity.²³ But more often the suggestion is that I am complicit in colonialism if I let myself enjoy the kindly old dramatist. But that is to contend that I am really seeing the duck when I think I am seeing the rabbit: I am being drawn to accept the enslavement of Native Americans when I think I am considering the nature of theater. Surely it is my intellectual task to make sure I do not confuse the two readings, that I do not let the kindly old dramatist justify the colonist. Achieving that kind of nuanced thinking is one of my goals as both critical reader and world citizen.

After all, if I am correct, understanding The Tempest is not to be equated with seeing either the colonialist or the metatheatrical allegory. Rather, seeing and being able to pursue both of them is how I develop my understanding, and being able to explain them to others is one criterion for having that understanding. I do believe that I have a more comprehensive grasp of the play for looking at it from Caliban's point of view and for noticing parallels between the play and the colonial enterprise of Early Modern England, but I can gain that benefit without repudiating allegories like the metatheatrical reading.

Each perspective illuminates certain aspects of Prospero: what is an insignificant spot on the back of the duck's head becomes the rabbit's mouth. Thus the colonial interpretation makes me notice Prospero's irascibility because that feeling is a natural accompaniment to a position of domination: in order to feel superior to the dominated class, the dominator tends to focus on perceived inferiorities that also irritate him. This irritation actually contributes to feeling little guilt at being the oppressor. In short, the irascibility becomes a natural part of Prospero, not a somewhat puzzling eccentricity, perhaps left over from the magician in the commedia dell'arte.²⁴ Power, including the power of the colonialist, is morally dangerous, as is suggested by the idea that the exercise even of white magic is potentially corrupting. Prospero is a kindly man for whom cruelty has become a habitual way of behaving.

Yet a metatheatrical reading of Prospero makes me notice aspects of him that do not fit with an easy condemnation of the colonial master. After all, he is genuinely affectionate toward Ariel, he seems unsurprised and unshaken in his love for Miranda by her defending Ferdinand and her disobedience, and above all he renounces his magical power at the end of the play. Of course these facts that the metatheatrical reading highlights are left over to be dealt with in the colonialist reading, as is the impulse they give toward a less hostile judgment of Prospero. The paradox is that he is the chief giver of freedom in a play that glorifies freedom, but he can play that role only because he is also the chief enslaver, as is vividly dramatized in the enslavement of Ariel and Caliban and the mock-enslavement of Ferdinand.

Prospero is a complex character—irascible, manipulative, occasionally forgetful—yet to most readers and viewers he is on the whole likable, even admirable, especially because of the admirable qualities I have pointed to and because his project seems aimed at benefiting everyone. That likableness is a dilemma for the colonialist reading. One option is to condemn the play as an apology for colonialism and patriarchy: it manipulates us as Prospero manipulates the other characters, but not for our own good: we learn to accept domination and hierarchy, especially over those perceived as different. Presumably the defenders of this theory exempt themselves from the supposed ill effect of the play; they are able to see from Caliban's perspective despite Shakespeare. I am suspicious of such readings; they seem like an inverted version of the old view of Shakespeare as providing sugarcoating for the masses and a deep and very different meaning for the cognoscenti,²⁵ though of course these critics suggest that Shakespeare's heart is in the sugarcoating, while the deep meaning and keen vision come from the critic.

A more fruitful approach in the first instance is to try to incorporate a partly favorable reading of Prospero into the colonialist interpretation. And surely, for a modern reader at least, one implication is that it does not take terrible people to do terrible things. It is the world of melodrama in which all colonialists and enslavers are monsters. But even in the most anticolonialist reading The Tempest is a fable of colonialism, not a realistic map of the process. Given that distorting mirror of fantasy, the differences can be as significant as the similarities. The whole theme of renunciation in the play, including Prospero's aim of emancipation, by which he gives up his magic power, creates a sharp contrast with both the Roman imperialism of the Vergilian analogy and the enterprise of New World colonization. If the adventurers and planters had been more like Prospero, the history of the New World might have been very different.

Having added these thoughts to my earlier cogitations on The Tempest, do I now understand the play? I can hope that my comments have persuaded my readers that I do, that they provide an adequate criterion for my understanding; but I feel sure that many of them consider me wildly eccentric: captive to crackpot theories or politically retrogressive or obsessed with the mechanics of response as opposed to the play itself or pedestrian and literal-minded. However, my primary concern in this essay is to give an accurate picture of the process of attaining meaning in Shakespeare's play. Responding to different interpretations, testing them against my detailed responses to the play and against each other, expecting to find insights in all sorts of odd places: all these are constituent parts of a necessarily sloppy and open-ended process.

Above all I seek to overthrow a traditional picture of dramatic meaning that fosters confusion and misunderstanding. There are several implications of rejecting the old picture in favor of the process I describe. There is no shining goal, the meaning of the play, that I might try to incarnate in my words. The concept of unity is only a heuristic tool, not the standard for accepting or rejecting my various thoughts about the play. I can draw insights from radically opposed interpretations, and I can ignore the most ingenious and cleverly argued constructions of ideas that I cannot connect with my own perceptions of the play. I should look carefully at my own feelings and intuitions, and I should listen sensitively to my voice (and others' voices) reading the play's words aloud. I can only test myself and improve my grasp of The Tempest by engaging in all of these activities, in the whole process of critical discourse. When I—and you—write criticism or teach classes or just talk about Shakespeare, we are using the best tools we have for climbing our individual mountains of understanding, and we will do it better for knowing what the process is really like.

Notes

1. Robert B. Pierce, “‘Very like a Whale’: Scepticism and Seeing in The Tempest,” Shakespeare Survey, 38 (1985), 167-73.

2. In addition to the texts listed below in nn. 4 to 8, some of the notable readings in this vein are O. Mannoni, Prospero and Caliban: The Psychology of Colonization, tr. Pamela Powesland (New York, 1956); Leslie Fiedler, The Stranger in Shakespeare (New York, 1972); Paul Brown, “This Thing of Darkness I Acknowledge Mine,” Political Shakespeare, ed. Jonathan Dollimore and Alan Sinfield, 2nd ed. (Ithaca, N.Y., 1994), pp. 48-71; Francis Barker and Peter Hulme, “Nymphs and Reapers Heavily Vanish: The Discursive Contexts of The Tempest,” in Alternative Shakespeares, ed. John Drakakis, 2 vols. (London, 1985), 1:191-205; Karen Flagstad, “‘Making this Place Paradise’: Prospero and the Problem of Caliban in The Tempest,” Shakespeare Studies, 18 (1986), 205-33; Thomas Cartelli, “Prospero in Africa: The Tempest as Colonial Text and Pretext,” in Shakespeare Reproduced: The Text in History and Ideology, ed. Jean E. Howard and Marion F. O'Connor (New York, 1987), pp. 99-115; Stephen Orgel, “Shakespeare and the Cannibals,” in Cannibals, Witches, and Divorce: Estranging the Renaissance, ed. Marjorie Garber (Baltimore, 1987), pp. 40-66; Meredith Anne Skura, “Discourse and the Individual: The Case of Colonialism in The Tempest,” Shakespeare Quarterly, 40 (1989), 42-69; Jeffrey Knapp, An Empire Nowhere: England, America, and Literature from “Utopia” to “The Tempest” (Berkeley, 1992); Jeffrey L. Hantman, “Caliban's Own Voice: American Indian Views of the Other in Colonial Virginia,” New Literary History, 23 (1992), 69-81; Richard Halpern, “‘The Picture of Nobody’: White Cannibalism in The Tempest,” in The Production of English Renaissance Culture, ed. David Lee Miller, Sharon O'Dair, and Harold Weber (Ithaca, N.Y., 1994), pp. 262-92; Jean-Marie Maguin, “The Tempest and Cultural Exchange,” Shakespeare Survey, 48 (1995), 147-54; Jonathan Bate, “Caliban and Ariel Write Back,” Shakespeare Survey, 48 (1995), 155-62; Jonathan Baldo, “Exporting Oblivion in The Tempest,” Modern Language Quarterly, 56 (1995), 111-44. See also the invaluable book by Alden T. Vaughan and Virginia Mason Vaughan, Shakespeare's Caliban: A Cultural History (Cambridge, 1991).

3. For the idea see Ludwig Wittgenstein, Philosophical Investigations, tr. G. E. M. Anscombe, 3rd ed. (New York, 1967).

4. Theodore Spencer, Shakespeare and the Nature of Man (New York, 1961), p. 200.

5. James P. Driscoll, Identity in Shakespearean Drama (Lewisburg, Pa., 1983), p. 155.

6. Stephen Orgel, “Prospero's Wife,” Representations, 8 (Fall 1984), 4.

7. Stephen Greenblatt, “Learning to Curse: Linguistic Colonialism in The Tempest,” in William Shakespeare's The Tempest, ed. Harold Bloom (New York, 1988), p. 67.

8. Lorie Jerrell Leininger, “The Miranda Trap: Sexism and Racism in Shakespeare's Tempest,” in The Woman's Part: Feminist Criticism of Shakespeare, ed. Carolyn Ruth Swift Lenz, Gayle Greene, and Carol Thomas Neely (Urbana, Ill., 1980), p. 289.

9. Student paper in Shakespeare course at Oberlin College.

10. More precisely this is one goal of reading and discussing the play. Certainly critical commentary, including the essays I discuss, can have other aims instead or as well, such as exploring some element of early seventeenth-century England, tracing the development of some social phenomenon such as colonialism, or even testing the validity of some intellectual tool. Still, understanding the play seems to me our central activity as teachers and scholars of literature, and a prerequisite to carrying out most of the others well. At any rate, my enterprise here is to describe the place of critical commentary in understanding the play. As a result I am not evaluating the whole intellectual reach of the essays.

11. For the vexed issue of what Wittgenstein means by “criterion,” see the discussions in Criteria, ed. John V. Canfield (New York, 1986), vol. 7 of The Philosophy of Wittgenstein.

12. I shall not try to defend my position that there are true and false statements of all sorts, with very different criteria of truthfulness. Thus I have no doubt of the truth of the statement that Miranda is a character in The Tempest, and not in The Winter's Tale. I shall be suggesting that even statements like “The Tempest exposes the patriarchy and colonialism behind Prospero's apparent benevolence” can be evaluated by their own criteria.

13. Epilogue, line 13. All quotations from the play are from The Complete Works of Shakespeare, ed. David Bevington, updated 4th ed. (New York, 1997).

14. I do not mean to deny Wittgenstein's demonstration that “games” is a family-resemblance concept with no single defining essence. See his Philosophical Investigations, 1:66-71.

15. For the term “click” and an exploration of this criterion for aesthetic understanding, see Ludwig Wittgenstein, Lectures & Conversations on Aesthetics, Psychology and Religious Belief, ed. Cyril Barrett (Berkeley, 1966), p. 19.

16. This is Wittgenstein's doctrine of meaning as use, developed at length in Philosophical Investigations. “For a large class of cases—though not for all—in which we employ the word ‘meaning’ it can be defined thus: the meaning of a word is its use in the language” (Philosophical Investigations, 1:43).

17. Thus Stephen Orgel in the Introduction to his invaluable edition asserts, “All interpretations are essentially arbitrary” (Stephen Orgel, “Introduction,” The Tempest [Oxford, 1994], p. 12).

18. E. M. W. Tillyard emphasizes regeneration as the completion of the tragic pattern in his Shakespeare's Last Plays (London, 1938). Cf. also Northrop Frye, A Natural Perspective: The Development of Shakespearean Comedy and Romance (New York, 1965), especially “The Return from the Sea,” pp. 118-39.

19. Barker and Hulme resist the term “theme” as implying an organic unity that they deny in The Tempest (“Nymphs and Reapers Heavily Vanish,” p. 197). Perhaps it is helpful to think of “theme” by analogy with the musical sense rather than as the key that opens a lock. For a subtle attempt to find organic unity in an interplay of themes, see Reuben Brower, “The Mirror of Analogy: ‘The Tempest,’” in his The Fields of Light (Oxford, 1962), pp. 95-122.

20. A minority deny it on one ground or another. See, for example, Anthony B. Dawson, “Tempest in a Teapot: Critics, Evaluation, Ideology,” in “Bad” Shakespeare, ed. Maurice Charney (Rutherford, N.J., 1988), pp. 61-73; Deborah Willis, “Shakespeare's Tempest and the Discourse of Colonialism,” SEL, 29 (1989), 277-89; John Hunt, “Prospero's Empty Grasp,” Shakespeare Studies, 22 (1994), 277-313; William M. Hamlin, “Men of Inde: Renaissance Ethnography and The Tempest,” Shakespeare Studies, 22 (1994), 15-44; Ben Ross Schneider, Jr., “‘Are We Being Historical Yet’: Colonialist Interpretations of Shakespeare's The Tempest,” Shakespeare Studies, 23 (1995), 120-45; and Richard Wilson, “Voyage to Tunis: New History and the Old World of The Tempest,” ELH, 64 (1997), 333-57.

21. Mark Van Doren, Shakespeare (Garden City, N.Y., 1939), p. 281.

22. See Wittgenstein, Philosophical Investigations, 2:xi, for the figure, a line drawing that can be read as either a duck or a rabbit. Wittgenstein uses the drawing to illustrate the concept of aspect, seeing something as something. E. H. Gombrich discusses the phenomenon in visual perception in his Art and Illusion: A Study in the Psychology of Pictorial Representation (Princeton, 1969).

23. The cliché is that to be unpolitical is to take a political stance in favor of the status quo. But surely the plausible form of that maxim asserts only that one should sometimes be political, not that one should always be.

24. See The Tempest, ed. Frank Kermode (London, 1994), p. lxxv, for puzzlement, and the discussion by Barker and Hulme, “Nymphs and Reapers Heavily Vanish,” pp. 202-3, for an explanation similar to mine.

25. See, for example, Harold Goddard, who uses this ploy in his The Meaning of Shakespeare, 2 vols. (Chicago, 1951), and justifies it in his preface (p. ix): “The purpose of the poet in this sense is often in direct contradiction with that of the playwright. It may even lead him in the interest of truth to distill ‘Sweet, sweet, sweet poison for the age's tooth,’ a line which, for our present understanding of him, may be the most important one in all Shakespeare's works.” For a critique of this two-audience view, see Richard Levin, “The Two-Audience Theory of English Renaissance Drama,” Shakespeare Survey, 18 (1986), 251-75.