Most literary criticism remains highly Romantic, not least that which sets out to be anti-Romantic. The underlying suppositions are not often argued or even stated. Indeed, it is the fact of their being taken for granted that renders them both significant and dangerous. We habitually oppose the individual genius to the merely conventional writer and celebrate the innovative over the traditional. Among the relatively modern formulations of the antinomy are the ironic, the ambiguous, and the paradoxical which are privileged over the straightforward. The code-breaker has precedence over the conformer, the scriptible over the lisible, and so forth.
The metaphysical pitfalls that surround this faith in what still passes for Modernism are evident on occasion: for instance, when Culler remarks that only literature of the sort associated with the first of these terms allows “expansion of self.”  Whatever this phrase may mean, it illustrates the predatory hunger of critical ingenuity, always ready to pounce on the flanks of the preceding beast before it has had a chance to enjoy its own repast. Thus Paul de Man in his lecture “Lyric and Modernity” evokes the value-laden polarity of the automatized and the defamiliarized (for him between a slavish mimesis and an autonomous obscurity) only to note slyly that “in theory the question of modernity could not therefore be asked of any literature of any time, contemporaneous or not.”  For him the categories are finally rhetorical and heuristic. By the lecture’s end, they have been quietly exploded.
The question of lyric modernity is most radically pursued by interrogating poems whose genre is generally considered to be formally and semantically conventional to an unusually high degree such as those medieval English texts gathered by Greene in his book of carols.  Twentieth century rereadings of a variety of medieval texts have made the discovery of their “modernity” no surprise, perhaps, but what is more significant is that poems such as these carols prove to be “modern” not in spite of their conventions, but through them. Their intertextual connections generate a greater semantic density and precision while, at the same time, creating a radically unstable ambiguity. The opposition between the conventional and the unconventional, between the referential and the formal as an index to modernity, or to literary value, is misleaning as practical criticism and mistaken as theory.
The misconception is reinforced rather than questioned by the reconstruction of a selected revisionist canon of the past, appropriating certain works only as approved in terms of the current critical vocabulary. Rather my present object is to examine works that few would regard as outstanding in order to demonstrate something of what constitutes literariness, no less for the carols’ original audiences than for ourselves. What seems at first glance to be “merely” conventional is often ain fact a play of signifiers which, whatever else it may be, represents an end in itself, a formal value intrinsic to all literature, just as certain musical forms delight the listener by elaborating recognizable variations, inversions, and reversals of topoi or themes or melody.
My reading of some carols points to just such verbal calisthenics,  but it does not collapse literature’s claim to referentiality. Far from being a wholly self-contained system, verbal art, when clearly distinguished from other forms of semantic data, retains a conflicted but inevitable reference to lived experience. This is true of religious texts no less than secular ones, though the former, by invoking the original “transcendent signified,” God, make a gesture that might seem to banish uncertainty. 
It is manifestly true that all language, and thus all literary uses of language, is intertextual and every text a revision of earlier texts. The levels of revisionary irony reach back indefinitely in a process graphically evident in strings of bathroom graffiti and at times in another of today’s lapidary genres, the automobile bumper sticker. There is thus no necessarily correct place to initiate investigation. To interpret Greene’s carol 448, a poem on the theme of the holly and the ivy, for instance, it would be legitimate to consider the use of these plants in folklore and ritual of pagan England and elsewhere. To clip the otherwise unlimited semiotic strain, I note only the obvious as an artificial point of origination: that the two plants, by remaining green and visibly alive at a time when many other plants wither to brown, naturally become associated with immortality and the promise of renewed fertility, and thus washed in the color of divinity.
These connotations were appropriated by Christians who linked the plants particularly with Christ and the Christmas holiday. This poem, though, reads like a courtly love lyric:
Grene growth the holy,
So doth the iue,
Thow winter blastys blow neuer so hye,
Grene growth the holy.
As the holy grouth grene
And neuer chaungyth hew,
So I am, euer hath bene,
Vnto my lady trew.
A[s] the holy grouth grene
With iue all alone
When flowerys cannot be sene,
And grenewode leuys be gone.
Now vnto my lady
Promyse to her I make
Frome all other only
To her I me betake.
Adew, myne owne lady,
Adew, my specyall,
Who hath my hart trewly,
Be suere, and euer shall. 
Here the symbolic force of the holly and the ivy may seem at first to be wholly comprehended in the cluster of meanings special/permanent/love. As such it is implicitly opposed to the flowers and trees so often surrounding other conventional love bowers, but which here are apparently dead. This meaning is so overdetermined that, in fact, the first distich of the burden is a proverbial expression meaning “forever.” 
The hard fact is, of course, that neither holly nor ivy nor human love is really eternal. There are internal revelations or, at any rate, hints of the rhetorical nature of that claim. Apart from the listener’s knowledge of the ultimate death of actual botanical specimens, the poet claims fidelity with the words “I am, euer hath bene,” yet promises in the third stanza to forsake the company of other women. His promise is itselkf the originating proposal for an exclusive relationship rather than evidence of its achievement. The longevity he praises is, at best, potential. Whether the whole idea is to be read as sincere or as a decoratively hyperbolic form of flirtation, the poem’s narrative movement directly contradicts the claim to stability as the speaker bids his lady “adew” in the last stanza.
The images of the holly and the ivy, which on the surface promised permanence, are further problematized by their intertextual relations. As Volosinov says, signs are without exception social, and exist in “interindividual territory,”  and the connections are especially insistent in the case of highly conventional texts. The other texts relevant to the poem’s interpretation, however, do not decode it in any simple way, providing unequivocal meanings to be inserted automatically once the key is known. Rather, as is typical of literary convention, interpretation becomes more complicated and polysemic as more texts are taken into account.
One willing to venture into the dubious realm of authorial intention would be rewarded to find that this poem is attributed to Henry VIII, but, without straying from the purely textual, the holly and the ivy are inescapably and deeply ambiguous. Among the other poems that Greene prints that use the motif of the holly and the ivy, the two plants are never equivalent or even harmoniously complementary. Among their common conventional significations is the male/female polarity apparent in “as the holy grouth grene,” but it more often appears in poetry (and in ritual) as a conflict or a tension than as a straight love story. The refrain of one carol says:
Nay, iuy, nay, hyt shal not be, iwys;
Let holy hafe the maystery, as the manner ys. 
This poem goes on to associate the holly with warmth, health, joy, and maleness, while the ivy is assigned cold, illness, sorrow, and femaleness.
In “here commys Holly, that is so gent”  holly has in fact gained the mastery called for in the previous poem to such an extent that the ivy has vanished altogether. What remains is a vaunt of holly from within the hall, still identified with maleness and power. This tendency is carried to the point of caricature in the version in Sharp’s English Folk Songs from the Southern Appalachians where the lovers have become a married couple, and the husband, finding “all my pleasure turned to woe,” cuts a switch of greenery with which to beat his wife into submission. 
Thus real stable domesticity transforms the lyric longing of the older text into ironic burlesque. The familiar Christmas carol “The Holy and the Ivy,” which, though not printed until the early 18th century, doubtless has roots in the Middle Ages, belongs to a different branch of the same tradition. Declaring that “the holly bears the crown,” it marshals all the machinery of supernatural divinity to bolster the claim. 
On the other hand, Greene prints three poems which celebrate the ivy and exclude the holly.  In the first two the ivy is identified with the semi-deified figure of Mary, and in the third it is praised for its more natural attributes as a medicine, an aid against soil erosion, and the like.
The meaning of the holly-ivy complex is thus clearly not reducible to specific substitutable terms. It is closer to the sort of structure Riffaterre calls a “geometry of extremes” and an “abstract dialectic.”  The polarities available from the handful of texts already surveyed would look something like this:
source of pleasure/source of pain
The terms are only partially unpacked by this schema, but the point should be clear that poetic convention has vastly thickened the plot, that the systems of signification between one poem and another serve to refine and sharpen the meaning. Poetic convention is a conceptual tool that enables the perception and expression of a continuously bifurcating systematization of reality. From the initial oppositions of nonexistence/existence, light/ darkness, water/land, by ever mores subtle gradation, the world is apprehended in all its complexity.
One function of presenting the world of love in highly symmetrical and patterned oppositions is simply to advance the intertextual series as a self-sustained play of signifying, in the spirit of the ecstasy (jouissance?) of intelligible but detached variation with the same sort of freedom from thematic harness that must inhere in the endless hallelujah of the heavenly chorus and in its earthly incarnation as a Bach fugue or a handsome porcelain teacup. Medieval commentary on the reputations of the troubadours and Minnesinger make it clear that improvisation within the bounds of convention was admired: the poet strove for originality by producing endless variations using the same generative formulae. The cognoscenti appreciated and cultivated this practice, enjoying the abstract play of form just as any musical phrase or prosodic pattern may be savored as an abstract pattern.  Indeed, it is only the ancient insistence, upheld in various forms yet today, that the study of literature is essentially a moral and ethical activity, that makes such a viewpoint sound frivolous or trivial.
A brief example from a very well-known lyric convention will perhaps clarify the point. Assume, rather arbitrarily, that the genesis of the “nature introduction” in poetry is the reverdie. It is easy to establish that, from the earliest date, a host of variations occur many of which are formal patterns, uninformative either about love or nature. The permutations proceed according to the same principle of bifurcation and continued division already observed. An elementary array of some common nature introduction possibilities (all of which occur) might look something like this:
Nature greens and I, too, am warmed by love.
Nature greens yet I suffer my lover’s coldness.
Nature grows cold and I, too, suffer my lover’s coldness.
Nature grows cold yet I am warmed by love.
Now, whatever else is happening in these poems, the poets are certainly playing games with the audience’s expectations. Whether the extrapolation of the audience is likely to make with the data initially given is fulfilled or frustrated or twisted in some manner, the same sort of game is enacted. The spirit of joy in free but patterned play that informs the fugue and the jazz solo alike is surely also one element constitutive of literature. Just as children arrange objects in groupings of hermetic meaning, artists construct artificial languages and audiences take pleasure in their own competence at following the formal action regardless of any referential content.
It is nonetheless true that, more clearly than in music or visual art, language always bears a relation, however mediated, to a non-textual reality of lived experience. The forms of patterning may depend on connections that arise prior to the encounter with the text. It is neither metaphysical nor willfully perverse to maintain that love contains latent aggression. Such phenomena as pornography and domestic violence are among the insistent testimonies to the reality of this particular paradox.
What is true of the male/female and love/hate polarities is no less true of the divine/human one. The notion of god is dependent on that of the mortal. Many hymns to god imply a hymn to secular love (as the only fitting model for the mystical experience), while, at the same time, insisting on an erasure of that limited terrestrial eros in favor of a greater. Every song of human love, likewise, is conditioned by the awareness of temporality, of the ephemeral nature of human life that makes pleasure at once more piquant and ultimately, as Ecclesiastes would have it, vain.
The balanced contradictions that can so richly convey dialectical information about the romantic lives of people can also express the same theme from the other side in the desire for the obliteration of all dualities and the yearning for a perfect (or at least perfectly simple) world. The oppositions are never wholly absent; they are frequently quite palpably absent by being explicitly said to be lacking. The explosion of opposites in religious discourse is a theme distributed world-wide , forming a poignant if inverted testament to the tensions present in the only life that people really know.
In the holly and the ivy poems that address religious themes, the polarities are evoked implicitly even if only one element, that is to say, one of the two plants, is mentioned. To listeners familiar with the pairing, absence is as significant as inclusion. In one of the Mary-ivy poems a mysterious lady who visits the speaker in bed turns out to be the Holy Virgin.  Here there can be no dilemma such as that presented to Sir Gawain in a similar situation – rather the scene represents perfect resolution of conflict. The poem’s persona is wholly passive. In an infantile way he receives the benefits of Mary’s presence. The real-life problematics of male/female relations associated with the arena of the bed only emphasize the transcendent otherness of this divine caller.
Another literary formulation of divine/human love is exemplified in the same poem. “Moder sche ys and maydyn trewe” is a cliché for a good reason. In the spirit of Tertullian’s famous dictum, the riddle-like paradox is highlighted and made the authenticating seal of a mystery rather than in any way undermining it. Real mothers, real virgins, and real lovers participate in the Virgin’s glory through their devotion while at the same time they are set well apart from it. In poem 261 the conventional theme of the lament of the abandoned woman who finds herself pregnant is placed in the mouth of Mary for whom, of course, the pregnancy is the occasion of unmitigated joy.  With a host of parallel texts in their minds, including other ivy poems, the audience can, in a way, share in the Virgin’s divine nature while appreciating the gap between her unique position and their own more worldly nature.
A final example of the fruitful intertwining of divine and human strands is the familiar Corpus Christi carol  in which the description of a courtly human setting is, by the concluding line, vaulted into the divine realm transforming an earlier vague sense of mystery into a great tremendum.
The function of convention, then, is not to repeat an automatized expression until is retains little meaning and no power, but rather to trope on it, to turn it, and one of the subtler twists may be to leave it as it is, since the context will unavoidably differentiate every instance. Conventions tend to appear in regularly patterned clusters. The antinomies in a simple holly/ivy text proved to be far from arbitrary. It is relevant to comment on the alienness of one sex to the other and the extraordinary dissolution of the difference in love, or on other themes suggested by the conventional manipulation of the holly and the ivy, but, in the end, if paraphrase at whatever length were adequate to reproduce the significance of the poem, poetry would be redundant. The meaning of the text, unlike all other (non-aesthetic) forms of discourse, is not susceptible to exhaustive paraphrase. This is not, however, due to its information lying in some metaphysical or esoteric realm “beyond words,” but, on the contrary, because of the peculiarly word-based, or literary, sort of message it brings.
This principle may have seemed a few decades ago very much a la mode, but the fact is that it forms the basis for the precepts of classical and medieval rhetorical treatises like those of Matthew of Vendome, Geoffrey of Vinsauf, and John of Garland. These documents of literary theory are sytill sometimes regarded (in the manner of Manly’s famous article  of several generations ago), as stereotyped and hackneyed formulae for sterile repetitive poetry. In fact their technical recommendations, especially the whole maligned system of tropes and figures, all foster an increase in semiotic density, a linguistic “thickening of the plot.” Far from cultivating willful obscurity, these devices move toward greater precision, an enrichment and sharpening of the poetic language leading toward inclusion of a greater amount of data in the coding. Through troping and other techniques, through twisting recognizable expressions, ideas, sentiments, and even narrative flow in unexpectable ways, they realized what Geoffrey called the “rejuvenation”  of language.
The patterning of literary conventions makes them intelligible just as paradigms of verb conjugations make small phonetic variation meaningful. These patterns, while exemplifying the literary quality of formal play as a self-fulfilling value, also transmit information about lived experience. Poetry is a unique technology of consciousness in that it never presents its data straight. This constant refraction underlies the ancient association of literature with lying and of the literary text with irony, ambiguity, and the like, as well as illuminating the claims of deconstruction.
In discussing the intractable problematics of human experience, literary texts evoke relevant oppositions, tensions, and contradictions. They may or may not privilege oone of the terms, but they always contain or imply both, sucggesting that the dance of life is a formal exuberance that can be read only throiugh a dialectical approach to the structure of semantic content.
Religious texts affirm the same facts from the far side of the margin of reality. There the opposing terms are adduced only to be shunted aside in the denial of duality or exacty balanced in a formula that encodes true Presence in encyclopedic inclusivity which either contains or banishes all antinomies. Apparently stereotyped linguistic usage is in fact the very freshest, because it is the most belated, a rehashing of old images toward new ends; it is always beyond and in reaction to the prior processed data; it is always hermeneia.
Speaking of Hermes or Thoth, the signifier god, the trickster (in the vernacular of black American street narrative, the signifying monkey) Derrida says, “This messenger-god is truly a god of the absolute passage between opposites. If he has any identity – but he is precisely the god of non-identity – he would be that coincidentia oppositorum to which we will have recourse again. In distinguishing himself from his opposite, Thoth also imitates it, becomes its sign and representative, obeys it and conforms to it, replaces it, by violence if need be . . .a sort of joker, a floating signifier, a wild card, one who puts play into play.” 
1. Jonathan Culler, Structuralist Poetics (London: Routledge & Kegan Paul, 1975), p. 129.
2. In Rueben A. Brower, Forms of Lyric (New York and London: Columbia University Press, 1970), p. 151.
3. R. L. Greene, The Early English Carols (Oxford: Clarendon Press, 1977).
4. The term is derived from Michael Riffaterre. See his Semiotics of Poetry (Bloomington: Indiana University Press, 1978), p. 78.
5. The question is raised by Jacques Derrida, Of Grammatology (Baltimore and London: Johns Hopkins University Press, 1976), p. 20.
6. Greene, p. 273.
7. V. N. Volosinov, Marxism and the Philosophy of Language (New York and London: Seminar, 1973), p. 12.
8. Greene, poem 136, p. 82.
9. Greene, poem 137, p. 83.
10. Cecil Sharp, English Folk Songs from the Southern Appalachians (London: Oxford University Press, 1952), p. 341.
11. Included in The Oxford Book of Carols (London: Oxford University Press, 1928), p. 80.
12. Greene, poems, 138, 139, and 139.1, pp. 83-4.
13. Riffaterre, pp. 44 and 46.
14. See, for instance, the mercurial, never-repeating stanza form of the well-known Harley lyric “When the nightegale singes.”
15. Examples of the praise of a primeval undifferentiation include Lao Tzu or even the Golden Age depicted in the Metamorphoses of Ovid or in Vergil’s Georgics.
16. Greene, p. 164.
17. Greene, p. 195.
18. Greene, poem 332, p. 195.
19. 19. John M. Manly, “Chaucer and the Rhetoricians,” Warton Lecture on English Poetry, Publications of the British Academy (London: Oxford University Press, 1926).
20. Ernest Gallo ed. And trans., The Poetria Nova and its Sources in Early Rhetorical Doctrine (The Hague: Mouton, 1971), line 769.
21. Jacques Derrida, Dissemination (Chicago: University of Chicago Press, 1981), p. 93.