Angela Esterhammer, Creating States: Studies in the Performative Language of John Milton and William Blake. Toronto: University of Toronto Press, 1994. xvii + 245 pp. $45.
In a book described as the concluding chapter of a decade long investigation of the aesthetic ideology, Jerome McGann argues that Blake is exemplary for radically resisting the “formal” and “organic principles of poetry and imagination” entrenched by Kant and Coleridge, principles unreflectively reproduced by a certain high romantic criticism ever since.1↤ 1 Jerome McGann, Towards a Literature of Knowledge[e] (Oxford: Clarendon P, 1989). Page numbers hereafter cited in the body of the review. Blake’s poems and designs are not, or not merely, “a dance of forms,” McGann insists, but “the textual ‘performances’ of his imaginative communications” (32); they are “deed[s] of language” (18) and “a set of actions carried out in the world” (4), whose “‘great task’” it is to effect “social and psychic overthrow” (25). Significantly, the critical rhetoric with which McGann brings out this begin page 30 | ↑ back to top resistance to the aesthetic ideology is drawn largely from speech-act theory. The point is that Blake’s texts do things with words, disruptive and self-consciously “gestural” and “performative” (12) things, and, as such, they “must be grasped [pace Habermas] as a type of communicative action.” Their “truth-experience” (7)—McGann does not say “truth”—lies entirely in the lively social transaction that they create, complicate, and to which they are contingently exposed. The year that McGann’s study was published also saw the appearance of Robert N. Essick’s Blake and the Language of Adam, a signally important book that, like McGann’s, evokes speech-act theory in the context of a broader hermeneutical argument. And like McGann, Essick shifts the emphasis from what Blake’s texts mean to their productive or illocutionary force within a social context. But it is there that the similarity ends. Far from exploiting the absences and differences inhabiting his own performances, as in McGann, Essick’s Blake strives for an ideal speech situation in which all semiotic things—sign, referent, recipient—are identified. For Essick, Blake certainly recognizes the differential and arbitrary nature of conventional signification, but remains committed to a radicalized literal expression patterned after the “kerygmatic or ‘performative’ gesture” of Christ’s blessing hand or Adam naming the beasts.2↤ 2 Robert Essick, William Blake and the Language of Adam (Oxford: Clarendon P, 1989). Page numbers hereafter cited in the body of the review. In speech-act terms, the complete uptake of Blake’s signs and the realization of their saving significance are at least theoretically possible within “the community of faithful recipients” (26). Under these conditions, “Jerusalem” would be what Alphonso Lingis calls, in a quite different context, a “city of communication maximally purged of noise,”3↤ 3 The Community of Those Who Have Nothing in Common (Bloomington and Indianapolis: Indiana UP, 1994) 86. the site of an absolutely felicitous performance in which the shared apocalyptic competencies of speaker and listener ensure the realization of meaningfulness without remainder. As a preface to his evocation of this perfected linguistic “State,” Essick reads Blake’s Adam Naming the Beasts, where he espies two mutually exclusive ways of doing things with words: on the one hand, the privileged performative that names the zoa into existence; on the other hand, the slippery coils and recoils of unmotivated language games that Essick identifies with “Nietzsche, Sartre, and Derrida” (16)—presumably the postmodern equivalent to Blake’s “Bacon, Newton, & Locke.”4↤ 4 David V. Erdman, ed., The Complete Poetry and Prose of William Blake, rev. ed. (Berkeley: U of California P, 1982) 218. Hereafter cited as E followed by the page number.
The fact that the principles of speech-act theory insinuate themselves into these quite divergent, if similarly strong readings of Blake says as much about the complex fate of that theory in contemporary criticism as it does about the artist’s linguistic practice. McGann’s and Essick’s studies form important critical pretexts—acknowledged as such—for Angela Esterhammer’s lucidly argued and elegantly written book, the first to take up the question of performativity in Blake’s work in a sustained and explicit fashion. As an inaugural study, the book can only introduce us to this very large subject, but what it does say is consistently illuminating and often provocative. (Esterhammer’s study is nominally about speech-acts in Milton and Blake, but I do not think that it is unfair to the book to claim that its focus and its most engaged negotiations lie with Blake.) For Esterhammer, speech-act principles offer a clarifying precision to the discussion of Blake’s texts, texts that forthrightly seek to create and recreate worlds with words, and that challenge (but, also, in some cases, reproduce) the forms of authoritative speech that police and produce the social body. That Blake was fascinated by the effectual nature of language and visionary art is of course nothing new. What is sharply original about Esterhammer’s study is its emphasis not only on the variety of performativities in Blake’s texts, but also on the ways in which the question of doing things with words goes to the heart of a number of related issues and problems: the constitution of the creative subject, the limits of aesthetic representation, the hermeneutics of prophetic discourse, and the poet’s negotiations with origins and originality.
Paradoxically enough, from a certain perspective speech act theory would seem to find unpromising ground in Blake. It is true that he creates texts that deliberately eschew constative statements—“forms of worship” (E 38), he would call them—preferring instead to conceive of his work as a performance “where meaning emerges in and through the encounter between reader and text” (175); but he is also the notorious builder of those “wall[s] of words”— to use De Luca’s memorable phrase5↤ 5 Vincent Arthur De Luca, Words of Eternity: Blake and the Poetics of the Sublime (Princeton: Princeton UP, 1991) 89. —whose “despotism in symbols”—as Coleridge once said6↤ 6 “To C.A. Tulk,” in Blake Records, ed. G.E. Bentley, Jr., (Oxford: Clarendon P, 1969) 252. —threatens the very basis of performativity; by short-circuiting what J. L. Austin calls “audience uptake,”7↤ 7 Cited by Sandy Petrey, Speech Acts and Literary Theory (New York and London: Routledge, 1990) 5. Blake’s utterances always run the risk of failing to do anything, as most of his contemporaries and many of my struggling students could well attest. Blake’s vision of words and deeds is almost always self-complicating. For example, the English visionary calls for freedom from the coercive force of conventional speech acts; yet he also recognizes that revolutionary subjects—epitomized by Orc—are actants who play roles in social and psychic scripts that can exceed and precede them. When Los’s song is described as “‘uttered with Hammer & Anvil’” begin page 31 | ↑ back to top (206), Blake’s alter ego clearly embodies the power of speech that has become act; yet Blake reserves his most vivid visual depiction of him in Jerusalem (plate 6) with his tools standing silent beneath the smothering presence of the Spectre. As an idealist, Blake can pattern his creative work after God’s fiats, even though as a more suspicious cultural critic he condemns the scurrilous ways in which “societal institutions take words like ‘God’ and ‘eternal,’ which in other contexts would convey transcendent authority and evoke a realm of language in which saying and doing are the same, and appropriate them into a logic of human laws and conventions” (157). Finally, as Esterhammer argues, Blake is conspicuously uncomfortable with the notion of an other-worldly authority speaking on behalf of humankind, and knows, like Nietzsche, that “cognition requires restriction and knowledge is never knowledge of a whole” (147); yet the redemptive plot of Jerusalem depends for its denouement—if that is what it is—on the outside intervention of the “Divine Voice” whose purely effectual utterances authoritatively announce an escape from the prison house of language.
Esterhammer begins her examination by in effect refining the hierarchized opposition between motivated and unmotivated language that underwrites Essick’s study. In its place, she works up a distinction between the “sociopolitical performative” and the “phenomenological performative,” each of which tends to generate its own mode of interpretation. Sociopolitical performatives explicitly obtain their illocutionary power from the cultural and historical circumstances of their utterance, and are associated with speech-acts made by figures within Blake’s texts. The moralizing declarations of the Priest in “A Little Boy Lost” are a case in point: here, normative terms like “holy” and “fiend” are much more than mere descriptions; they are disciplinary acts that subject the child, fitting him to the regulatory ideal of the theological regime. As Esterhammer demonstrates in close readings of selected passages from Songs of Experience, The Marriage of Heaven and Hell, The Book of Urizen, and Jerusalem, what drives Blake’s social work is the far-reaching “recognition that societal institutions only exist in so far as they are created by speech acts (charters, vows, declarations of independence) and kept in existence by the exercise of verbal performativity” (27). The phenomenological performative, on the other hand, grants a certain privilege to the speaker, and prompts an analysis that focuses on the way the utterance (re)produces the power and originality of the speaker or poet. The paradigm for this kind of creative and effectual language is the divine creation by the Word, the originary speech-act, as it were, that Blake mimics by creating illuminated texts that derive “authority from private visionary consciousness” (25). For this reason, Esterhammer argues, visionary poetry “in its entirety may be regarded as performative discourse in that it is a sustained act of asserting authority” (33) on the part of the poet. Where the function of sociopolitical performatives is to control and to disempower the listener, phenomenological performatives serve the purpose of constituting and empowering the speaker.
Esterhammer argues that Blake’s Bardic admonitions—like Milton’s “‘Mark well my words’” (E 100)—constitute both a construction of subjectivity and an assertion of the subject’s authority. Or in her neat paraphrase: “‘The one who is saying this is I, and I am the one authorized to say it’” (33). As the circular logic of this phrasing suggests, however, the visionary’s authority comes at a price, for it “risks being exposed as always and only a function of language” (33). This is a risk that Esterhammer acknowledges but perhaps underplays in her book, since she is clearly drawn—as are so many Blakeans—to the heroic image of Blake creating a world “in defiance of the existing one, to demonstrate the poet’s imaginative independence from the social conditions of his or her utterance” (25). From the perspective of contemporary language theory, this independence is an illusion, for performative acts are always at a certain level reenactments, the reiterative or citational effect of socially sanctioned practice. In as much as phenomenological performatives refer to utterances that are “non-conventional, extra-societal, deriving from the will or intentionality of the speaker alone” (13; emphasis mine), they are, strictly speaking, a contradiction in terms: an utterance spoken without reference to any social context would be what Derrida calls “the vocative absolute,”8↤ 8 Of Grammatology, trans. Gayatri Chakravorty Spivak (Baltimore: Johns Hopkins UP, 1978) 112. and unintelligible as such. Blake does not see performative language as always already sociopolitical, an argument that Esterhammer makes with such vigor that her position would at times appear to be indistinguishable from that of her subject. Consistently wary of contemporary readers making Blake over into their own (“post-structuralist”) image, Esterhammer insists that Blake, especially the vatic, inspired Blake, boldly lays claim to the presentist notion that he is the exclusive origin of what he says. Here, the uptake of prophetic meaningfulness is presumably more a matter of shared belief (or faith) than linguistic competency, and, as literary critics we do not talk about such things because they make Blake too dangerous by half. Perhaps we could say that the pure phenomenological performative functions not so much as an a achieved fact—we can say that it happens, but how would we know it had happened?—and more as a figure of visionary desire, no less powerful for being that. It is the image of linguistic perfection, whether we believe in it or not, that throws the contingent and often coercive nature of fallen speech-acts into sharp relief.
For Esterhammer, the sociopolitical performative and phenomenological performative interact and contradict each other by turns throughout Blake’s work. The artist begin page 32 | ↑ back to top comes by this dichotomy naturally, however, since, as Esterhammer shows, the same thing is to be found in Blake’s two great precursors, the Bible and Milton. In a manner reminiscent of some of the work of Kenneth Burke, Esterhammer reads Genesis as a pretext for both the argument and the rhetoric of modern language theory. In the “Priestly” or “P” myth of creation in Genesis 1, we of course witness the most vivid example of a vision of language “which can create things from nothing so that the resulting world is co-existent and perfectly correspondent with the words” (52). Against this spectacularly effectual but asocial (or pre-social) instance of doing/saying, Esterhammer contrasts the “Jahwist” or “J” text of Genesis 2 and 3, in which God’s prohibiting, cursing, and naming—speech-acts all of them—organize and regulate an already existent life-world through the power of authoritative utterance. Explicitly oriented towards a present and future disciplinary social context that is at once instituted and dominated by speech acts, the “J” text thus represents the inaugural example of the sociopolitical performative. Not unlike what Blake provides in The Marriage of Heaven and Hell, Esterhammer’s reading of Genesis amounts to a genealogy of morals that brings out how the Judeo-Christian culture naturalizes its socially circumscribed norms: in so far as God’s speech-acts “establish domination and subordination as the characteristic terms of relationships” between men and women, humans and animals, humans and the natural world, they provide a kind of theological alibi for the human, all too human, history of suffering, inequality, and “power relationships” (57) that follow.
The conflictual nature of the performative emerging from the Genesis account has important implications for texts claiming visionary or inspirational status. Esterhammer begins by pointing to tensions inhabiting some of Milton’s prose works as well as Paradise Lost, where his “desire to represent and imitate the phenomenological performative, as it appears in divine creation, is repeatedly threatened by the intrusion of the sociopolitical performative” (67). Interestingly, for Esterhammer this threat does not come from any explicit sense that all utterances are conventional, even and especially those that naturalize their authority by affecting to be purely “creative” or “expressive” in the manner of God’s fiats. That is a knowledge about performative language that must await Blake and romanticism, although why precisely the late eighteenth century would possess it where the late seventeenth did not is a question that could stand a little clarification in the book. Instead, the hazard that Milton confronts is displaced into a kind of hubristic embarrassment about presuming that one’s work is or can be truly self-originating. This is never more sharply legible than at those points where Milton reflects upon the origins and nature of his own visionary power, as in the invocations to Paradise Lost. Milton starts with explicit gestures towards firstness but finally figures himself “in the position of one who comes second, a revisor and reshaper of received material” (77). In Books 1 and 3, Milton reveals his desire for and faith in the performative ability of his language to summon a world of phenomena into existence, but by Book 7 this desire is overtaken by a counter-sense that creative utterances are also a matter of arbitrary, violent impositions and articulations spoken out of a limited position whose inspirational authority rests in large part with the consent of a community of believers—i.e., language acts that more closely resemble the conventionalized utterances of the sociopolitical world than the protected inward realm of the phenomenological performative. This crossing between what Milton desires and what he knows, or between what he says and what he does, represents an important—though underthought—moment in literary history, for here Milton takes “a first step toward implicating creative utterance in a structure of repetition” (89).
The same questions of authority, performativity, and representation pre-occupy Blake, whom Esterhammer characterizes as “much more conscious of, and therefore anxious about, the dichotomy between language which derives its creativity from individual will and language which wields by common consent” (64). Curiously, the Blake that actually emerges from Esterhammer’s discussion does not come across as particularly anxious, and, indeed, seems most often in command of his vision of words, coolly manipulating the performative dichotomy in which he is also said to be caught. Nevertheless, the ways in which the phenomenological and the sociopolitical performatives interact in Blake’s texts varies considerably over the course of the production of the illuminated works. Songs of Innocence and of Experience stage them as a non-dialectical opposition: “Innocence” names a condition of felicitious speech-acts in which the Child’s illocutionary force and perlocutionary effects are immediate and forceful; “Experience” is the realm in which speech acts either fail, as in the case of the Bard’s summons to Earth, or succeed only too well, as is the case with the social discourses that subjugate their auditors. Against the almost canonical position in Blake studies to interpret Songs of Innocence through the ironic lens of Songs of Experience, Esterhammer thus presses for a reading that “more fully” appreciates the significance of the “idealized scene of discourse” (130) being envisioned there.
I take Esterhammer’s point; part of recovering a more dangerous Blake may well lie in allowing him to have a vision of words wholly different from that of the late twentieth century. But the question remains why it is necessarily the case that the communicative action of the Bard is in effect ironized by the felicity of the Child, but not vice versa. In other words, aside from our firm persuasion that it is not so, why can’t the Child’s performative success be read as a dreamy projection about language rather than an ideal against which to measure the infelicities of spoken experience? I am not sure if this question can be dismissed simply begin page 33 | ↑ back to top as one emerging out of “a Derridean anxiety imposed onto the Blakean text” (129), especially since Esterhammer’s subsequent discussion of The Book of Urizen and The Marriage of Heaven and Hell fully demonstrates Blake’s extreme sensitivity to the fact that phenomenological performatives, with their claim to authority and efficacy, can instantly become the alibi for speech-acts of sociopolitical violence. (We might further ask what a phenomenological performative is and must be in the first place if it is supplementally open to such violent “appropriation” (157) by its demonized other? Does the phenomenological performative suffer this appropriation as a kind of accident, or does its susceptibility to corruption point to its always already being sociopolitical in nature? Is it possible that the sociopolitical realm produces the conceit of an appropriated or repressed “creativity” in the realm of the phenomenological in order to rationalize its own strategies?) Even if Blake held out the possibility of a truly ideal speech situation when he produced the Songs, the subsequent reprinting of that text alongside The Book of Urizen and The Marriage exposes it to a hermeneutics of suspicion that inevitably reframes its idealistic claims about an “innocent” performative language of individual vision.
The Marriage—which Esterhammer characterizes as the Blakean text most open to speech-act analysis—would seem to be a text that is framed by the linguistic problematic it frames, unwilling to dwell on either side of the performative dichotomy that organizes Songs. Through his parody of “the Bible’s various language games, including prophecy, wisdom literature, law, and history” (160), Blake denounces the authoritarian claims of those utterances that claim universality but in fact expressly perform the disciplinary labor of Church and State. Against the systematizing codes of the Priests, Blake opposes the creative language of the Poet; but he does so even though the Marriage’s pervasively dialogical form reminds us that all utterances are dependent upon context and perspective, even as every quantum of energy relies upon its circumference to make it legible. The title of the text nicely captures the problem: the marriage ceremony evokes the primary example of sociopolitical performativity in language theory, the exemplary instance of an utterance whose illocutionary power and perlocutionary effect rely upon a certain minimal agreement between speaker and audience; yet Blake makes his marital declaration, like all the other declarations in the poem, “utterly without the authority or the societal consensus that would give him the right to make such a pronouncement” (170). The (un)solemn union of “Heaven & Hell” asserted by the poem is thus not the effect of a collective accord, but an imposition—even an act of rhetorical violence—whose authority is produced only in and through its own performance. How to account for this curious contradiction in speech-act terms? Or as Geoffrey Hartman once asked, “Where does Blake get his authority from?” (216). Esterhammer implies that while intellectually Blake recognizes that his devilishly individualistic and non-conventional claims are made without any substantiating authority, emotionally he remains attracted to the self-grounding power of phenomenological performative, especially since the Marriage marks the “turning point” (172) in which Blake begins to lose “whatever interpretive community he was ever able to address” (173). At the conclusion of her discussion of the Marriage, Esterhammer recuperates the situation somewhat by suggesting that even without an audience to make his utterance felicitious, Blake “performs” a “marriage . . . in the writing of the poem itself, and we instantiate it in reading” (172). Yet her own suggestive reading of the poem rightly complicates what this instantiation could mean. Far from simply realizing the communicative action of Blake’s performative utterance, Esterhammer’s instantiation of Blake’s work in her own study helps to bring out how the problem of performative legitimacy is complexly symptomatic of a cleft inhabiting declarative discourse in general. As Derrida has shown with regard to declarations of independence, revolutionary assertions involve a strangely duplicitous twist in thinking: they create the very state that they must already be in in order to bring that state about.9↤ 9 See, for example, “Declarations of Independence,” trans. Thomas Keenan and Thomas Pepper, New Political Science 15 (1986): 7-15. Lacking any consensual authority, the force and efficacy of Blake’s visionary utterance is derived from its own declaration in the form of his text. Yet in order to have a vision and to make his pronouncements, Blake must already in some sense be visionary. Or in Geoffrey Bennington’s terms, “there is no performative which does not also involve an at least implicit description of the state of affairs it produces.”10↤ 10 Jacques Derrida (Chicago: U of Chicago P, 1993) 233. The authority of Blake’s declaration of independence from conventional speech-acts—again, not unlike all declarations—is thus scandalously unstable, since it is grounded in its own act of grounding. The circular and recursive logic of the illocutionary force of Blake’s poem is this: it must verbally effect what it requires to be effectual. As such, the performance always lags slightly behind itself, never a purely phenomenological utterance springing from the creative will because of the hidden way in which it derives its authority from an always earlier speech act extending “backwards” towards an irrecoverable “past.” To pick up on Esterhammer’s language: visionary speech is “a hybrid of the phenomenological and the sociopolitical performative” (177; emphasis mine), that is, a grafting of an act upon a description.
Esterhammer divides her concluding chapter on Jerusalem into two movements. Beginning with an analysis of different speech-act strategies adopted by figures within the text, she shows how those utterances provide a means by which to apprehend the visionary performance of Jerusalem begin page 34 | ↑ back to top as a whole. True to the teleological assumptions underwriting her account of Blake’s development as an artist, Esterhammer treats Blake’s epic as the culmination of a lifelong negotiation with performativity. It is here, in the act of “creating states,” that Blake finally finds a “compromise” to the problem of transcendent and convention-based ways of doing things with words. “Jerusalem” is the pre-eminent example of this settlement, a socio-political and visionary “State” that is said to be named—and framed?—into existence by the poem’s first and last words. Between these apocalyptic pronouncements, the narrative is noisy with unhappy speech acts—lamentations, exhortations, and curses—almost as if the true performative can emerge only when all the false have exhausted themselves. Albion’s sorrowful cry at the end of chapter 1 marks a critical moment in this progress: “What have I said? What have I done? O all-powerful Human Words!” (E 169). Albion’s apostrophe to signs registers his self-conscious awareness that language harbors an articulate life of its own, and that the authority of his words is thus not of his own doing. (Today, we would call this overwhelmingly mighty language discourse.) Esterhammer contrasts the agonistic and derived nature of Albion’s speech acts to the forcefulness of the “Divine Voice” that anticipates the redemptive outcome of Jerusalem and brings this outcome about in the form of the poem itself. The “Divine Voice” may be the direct expression of the creative will of the “Divine Family,” but the rhetoric and the strategies of its redemptive utterances are resolutely social. As an accommodation to the inhabitants of Albion’s Land, the “Divine Voice” awakens Albion by creating and recreating “States,” a word, Esterhammer astutely emphasizes, that possesses important political and communal resonances. Although Esterhammer does not describe it in these terms, a “State” possesses a formal rather than substantive integrity; as the sum of its speech-acts, it is an ongoing articulation and rearticulation of itself that has its “permanence” (as Blake puts it) in the continuity of this process. Blakean “States,” like nation states, are thus more or less provisional arrangements with the express purpose of “continually” staging a redeemed social universe. Esterhammer characterizes this felicitious outcome as a truce between the two forms of performativity, but it is perhaps more accurately a linguistic apocalypse in which the distinction no longer obtains because language itself has somehow been redeemed. It is precisely this transformation that Esterhammer sees being described (or perhaps even effected) in Jerusalem’s concluding plates.
Esterhammer identifies herself with those for whom Blake is interestingly resistant precisely because he is not postmodern, even if his insights illuminate and reproduce elements of contemporary language theory. In addition to its readings of individual texts, the strength of this book is that it opens up so many possibilities in other areas of Blake studies. For example, how do the principles of speech-act theory help us investigate Blake’s vexed representation of the feminine? At a recent Romantics conference, I heard Anne K. Mellor roundly declare that Blake “was sexist to the core.”11↤ 11 Mellor made this remark during the discussion period following the session on “Framing the Subject: Portraits and Frontispieces,” at the conference on Romanticism and the Ideologies of Genre, University of Western Ontario, London, Canada, 26 August 1993. but one of the things that speech-act theory has done—especially in the work of Judith Butler12↤ 12 See, for example, Bodies that Matter: On the Discursive Limits of “Sex” (New York and London: Routledge, 1993) 223-42. —is to demonstrate how no one is anything “to the core,” not even gendered. Speech act theory might well prove useful in moving us past the critical impasse created by Mellor’s curiously dismissive and preemptive statement. In critically discerning how ideological and gendered positions are produced and reproduced in performative acts of authoritative speech, we will perhaps understand more clearly what “sex” and “sexism” mean in Blake’s texts. In other words, speech-act theory encourages us to continue the task of reading—rather than dismissing—Blake as a complex speaker in a shifting cultural context. In this regard, Esterhammer’s book is exemplary. To be sure, Esterhammer offers us a melioristic interpretation of Blake, as perhaps befits a book dedicated to Northrop Frye. Yet she also emphasizes a Blake who was resolutely pragmatic, a Blake who might have said that although all speech is equal, some is more equal than others. Blake dispels the peculiar legal-constitutional hallucination that speech is free, and that what one says is entirely divorced from what one does. As Esterhammer concludes, “One could say that the aim of Blake’s art, like that of How To Do Things with Words, is to demonstrate the illocutionary force of constative statements, or to emphasize that all speech acts” (207). Speech does not simply happen; it is made by individuals in specific social contexts, in which some speak with authority, and thus with real consequences for those who are compelled to listen. The fractious and strangled cries that resound through Blake’s texts repeatedly show and tell us two things: words are never only words, and those who are not in a position to speak effectually are kept in that position of inequality by those who are.13↤ 13 I am thinking here, of course, of the argument that Catharine MacKinnon makes in Only Words (Cambridge, MA.: Harvard UP, 1993) 71-110.