In 1983 Jerome McGann assured readers of The Romantic Ideology that Wordsworth's greatest poems, 'like his position in the Romantic Movement—are normative and, in every sense, exemplary'.  'Exemplary', he explained, because truest to their time: 'such works transcend their age and speak to alien cultures because they are so completely true to themselves, because they are time and place specific' (McGann 1983, 2). For the previous decade M. H. Abrams had made the same assertion: 'Wordsworth ... was the great and exemplary poet of the age'.  The similarity of these claims suggests a continuity seldom emphasised: McGann's evading the question of the literary value of the canon has long tended to obscure the fact that The Romantic Ideology remained nevertheless an explicit defence of the traditional 'Romantic' canon—Wordsworth as central figure—as an appropriate site for critical and pedagogic activity. This may seem obvious, but it is necessary to stress the point because 'New Historicism', the (supposedly coherent) critical movement with which McGann's name is associated, has frequently been represented as hostile to the received canon of 'literature': that is, hostile not just to ideas found in that canon, but to the canon—perhaps any notion of 'canon'—itself. Such an assumption has often misled discussions of 'Romanticism' over the past decade and a half, obscuring other canon-related issues that have grown in importance.
The theory that 'New Historicism' is opposed to the 'Romantic' canon (or, indeed, other canons) quickly runs into evidential problems that are then perversely attributed to 'New Historicism' itself. To Jonathan Bate, writing in 1991, it was a 'paradox' that 'New Historicism' was continuing to concern itself with 'canonical texts' when theoretically it granted no greater value to those than to any other form of documentary evidence.  Thomas McFarland, perhaps the foremost anti-'New Historicist' critic in the field of '80s and '90s 'Romantic' studies, has variously supported the argument, but reveals its problematic status when he writes: 'deconstruction does not theoretically restrict itself to literary art, but regards any text whatever as a proper subject for deconstructive analysis... What is true of deconstruction is true also of its popular offshoot, New Historicism'.  The point would only be useful if it could be shown that other types of criticism were, somehow, 'theoretically restricted', but this McFarland does not, and presumably cannot do. Indeed his own form of 'qualitative' criticism (McFarland 26) is necessarily unrestricted at a theoretical level, or it would bewilder itself with its inability to demonstrate that a laundry bill was not a good poem.  Setting specifically 'Romantic' period criticism aside for the moment, though, it is clear that Bate's and McFarland's accusations of paradox are far from untypical. In a well-informed defence of older values, Josephine M. Guy and Ian Small, for example, refer several times to the de-privileging of 'literary' texts by 'New Historicists' before finally admitting: 'However ... new historicists have tended to focus their attention on the evidence provided by those works enshrined in the canon of traditional literary history: either on the work of Shakespeare ... or the Romantics ...'. They then count this as evidence for the 'illogicality' of the premises of 'New Historicism' and a few pages later describe it as an 'irony'.  The persistence of such assertions must mean that the theory and practice of 'New Historicism' are indeed at odds, or that 'New Historicism' has become an increasingly useless umbrella term for a body of criticism that is by no means cohering, or, simply, that 'New Historicism' has not properly clarified its relationship to the canon, or that it has done so, but not been understood.
One way to assess the problem might be to examine the critical genetics of 'New Historicism', particularly that which we encounter as (loosely) 'Romanticists'.  After all, McGann's central claim in The Romantic Ideology that critics of 'Romanticism' were too sympathetic to the series of evasive moves that allowed something called 'Romantic' to be written in the first place was not exactly new. De Man had called attention to it almost two decades earlier in the context of a discussion of Wordsworth.  Moreover the sense of novelty derived from McGann's combining De Man with something much older still: in both the 1980s and the 1810s and 20s the debate over the 'Romantic ideology' was a consideration of 'the issues between Hazlitt and Coleridge as to which was the appropriate totality that defined Wordsworth's poetry: social or metaphysical,'  his poetry seeming central to the argument about the political commitments of 'Romantic' literature. The Romantic Ideology was a work, in other words, very much in the tradition of Hazlitt's Spirit of the Age, at least as that volume had been generally understood. (Increasing interest in Hazlitt over the last decade or two is no doubt intimately linked to the rise to critical prominence of 'New Historicism'.) McGann agreed with Hazlitt that one could talk about a 'spirit of the age' and, like Hazlitt, he defined it with scant respect for any existing critical consensus. Neither Wordsworth nor his admirers in the 1820s can have been pleased to read his genius described as 'a pure emanation of the Spirit of the Age', only to have that 'Spirit' summed up as 'revolutionary' and 'levelling'.  Similarly, the Wordsworthian old guard of 'Romanticists' resented McGann's Marxist-orientated appropriation of Wordsworth in the 1980s. To acknowledge the parallel, however, is to recognise that McGann's 'spirit of the age' is as problematic as Hazlitt's has been revealed to be. Marilyn Butler, for one, has described Hazlitt's comments on Wordsworth as 'impish' and 'impudent', and James Chandler has shown how problematic they are when cited as a justification for the traditional Wordsworth-centred view of the period.  Nevertheless, it is a paradoxical consequence of the prevailing division of 'Romantic' period studies along political lines that Chandler's essay, published in 1990, was squarely aimed at M. H. Abrams, not, bar a very brief hint, at McGann and the 1980s 'New Historicists' with whom Chandler himself has often been associated.  Chandler might have done better to question how his demonstration that the 'spirit of the age' was a 'Romantic' construct questions a basic tenet of The Romantic Ideology and much of the criticism it empowered. Although that criticism has often been read as hostile to Abrams and the old guard of 'Romanticists'—not least by themselves—it continued to champion the idea of the 'representative man', or, more particularly, the 'representative consciousness'. This in turn preserved the idea of a 'Romantic' canon of 'representative' works. As the discipline starts dividing along new lines the affinities between the 'New Historicists' and their putative opponents may ultimately come to seem more striking than the differences.
Any critical construction of a 'spirit of the age' faces the problems evoked by Bakhtin in his warning description of 'ideological monologism':
Even when one is dealing with a collective, with a multiplicity of creating forces, unity is nevertheless illustrated through the image of a single consciousness: the spirit of a people, the spirit of history, and so forth. Everything capable of meaning can be gathered together in one consciousness and subordinated to a unified accent; whatever does not admit to such a reduction is accidental and unessential. 
Historically-orientated 'Romanticists' have differed importantly in regard to this prescription, and that difference is illustrated here by comparing late 1980s essays by Marjorie Levinson and Marilyn Butler. In 1989 Levinson and Butler, with McGann and Paul Hamilton, contributed reflections on the changing shape of period criticism to Blackwell's Rethinking Historicism. Levinson's, 'The New Historicism: Back to the Future', begins by setting out the paradox that widespread scholarly interest in our construction of the past has tended to blind us to the way that the past has constructed us, more particularly that 'this interest of ours ... might ... be an effect of the past which we study' (21). She then, at length, characterises the 'philosophical crux' that she considers common to 'Romanticism', historicism and our blindness to the effects of the past (25). Marx is the hero who cut the Gordian knot: 'Marx's great contribution to the historicist discourse was his demonstration that the subjectivity of the present, like that of the past, is objective; and, that this objectivity is at once absolute and historically, materially determined' (31). He can also free us from our current critical impasse: 'In our reluctance to ... yield up our own subjectivity to a critique by the past, we have rejected that moment in the Marxian argument which would put us outside the hermeneutic circle precisely by situating us within it' (34). In the second half of her essay Levinson offers a brilliant analysis of Wordsworth's sonnet, 'The world is too much with us' (1802-4), with a view to showing how its 'contradictions ... which I take as epochally symptomatic, do not anticipate the Marxian solution, [but] precipitate it' (47). 'In [Wordsworth's] elegiac and, thus, self-serving critique, we recognise the very form of our literary criticism, which redeems to murder' (49). In her conclusion she broadens this assertion into a general claim for why 'Romanticism' seems so important for us just now (specifically, of course, the late 1980s).
Marilyn Butler's much shorter article, 'Repossessing the Past: the Case for an Open Literary History', begins by asserting that although there are more readers of English worldwide than ever before, they are increasingly unlikely to have a special interest in British literary history. The fact that the traditional 'Romantic' canon seems obsolete is both cause and effect of this process. We are facing 'the unfortunate intellectual consequences of letting a small set of survivors, largely accidentally arrived at, dictate the model many of us seem to work with, of a timeless, desocialized, ahistorical literary community' (72). In the second part of her essay Butler examines Southey's Thalaba the Destroyer (1801), which she suggests demands a new kind of reading. She argues that the poem displays 'Southey's extraordinary intertextual range, his vulgarity in the strict senses that he's both thoroughly absorbed with popular art, and himself the medium by which traditional stories make their way down into the cultural water-table' (74). In this way, Southey's poem, 'serious and intelligent without ... being private and academic' (82), seems more relevant than works in the traditional canon to today's international and multicultural concerns. Also, by placing canonical literature in a new light, works like Thalaba may give that literature something of a new lease of life (Butler avoids the questions of how long this lease might be, and, by extension, what status the works of the traditional canon might eventually be expected to have). Butler concludes by calling for 'A new historicism, newer and more open than most work attracting that description at present' (84).
The concern here is not with the specific details of the arguments that Levinson and Butler advance, but with the broad nature of their arguments, the critical method they employ, and what this says about historical criticism, the idea of the 'representative' consciousness, and the 'Romantic' canon. It is clear that contradictory claims are being made: Levinson maintains that we have been chosen by the (specifically 'Romantic') past; Butler that we can, and should, choose a new past. Levinson sees a critical and philosophical problem; Butler an institutional and pedagogic problem. Levinson finds an answer, or a guide to the answer, in a canonical Wordsworth poem; Butler in self-consciously turning to non-canonical period literature. It is perhaps ironic, certainly striking, that Butler, British, assumes that the traditional, Wordsworth-centred 'Romantic' canon must fade, in interest and relevance, with the last vestiges of the British Empire, just as Levinson, American, asserts that Wordsworth has never been more interesting and relevant. And this can be pressed further: in justifying a detailed examination of a single, canonical sonnet on the grounds that it is 'epochally symptomatic', Levinson makes the type of critical gesture (whatever its theoretical inflexion) to which Butler is precisely opposed. McFarland, the champion of a knowingly 'Romantic' criticism, though hostile to both Levinson and Butler, would have to admit that Levinson's working method here has more in common with his own than with Butler's. Indeed Butler presumably had Levinson (among others) in her sights when she appealed for a 'newer and more open' 'New Historicism'. Within Levinson's criticism, however, it is not clear that such 'openness' would be possible, certainly not along such simplistic lines. No doubt she could write very well—and very differently from Butler—on Southey's 'vulgarity', but it would not be with a view to constructing the sort of dialogic, Bakhtinian view of the period that Butler wants. And Levinson's main reason for not writing on Southey would probably be that Keats is a more relevant model of the 'vulgar' early-nineteenth century poet: that is, in so far as we think we know him better, we can learn more from him. The canonical authors force the questions. 
The crux of the matter concerns Bakhtin's 'one consciousness'. Levinson finds it in 'representative' Wordsworth and so understands history in terms of its felt, mediated dialogue with us. Butler resists the 'unified accent', or at least admits conflicting forces into her unity (high v. low culture, male v. female, conservative v. radical writers, etc.), and so understands history 'objectively' as the sum of preserved discourse. Pace many of the critics of 'New Historicism', then, it seems clear that for Levinson great literature is not in any sense placed on a par with other parts of the documentary record as 'historical evidence', but rather is elevated into the most valuable and relevant form of 'history'—the representation of the 'representative' consciousness.  Butler's criticism, on the other hand, cannot be defended against such a charge at a theoretical level, though in practice—which is what is important—it privileges the complex historical engagements of specifically 'literary' writing. Nevertheless, it is Butler's, not Levinson's, criticism that justifies expansion of the canon, if not a wholly anti-canonical program. It is not unreasonable, indeed, to consider Butler the presiding genius of the massive reprint and recovery operation set in motion in the last two decades. Perplexingly, however, it is Levinson, through her association with McGann and The Romantic Ideology, whose name has been more consistently associated with a new historical threat to the 'Romantic' canon. The only reasonable explanation of this is that McGann's and Levinson's concern to transvalue works in the 'Romantic' canon was misunderstood as a project to devalue them. Hence Butler and other critics who simply went on distributing value over a broader range of texts actually created less controversy.
It should be clear, then, that two distinct types of historical criticism were developed in the 1980s, only one of which was hostile to the 'Romantic' canon. Moreover as 'New Historicism' better describes the other, there is no reason to find anything paradoxical or suspect in the allegiance of 'New Historicists' to canonical literature. It can be argued, further, that this allegiance to the canon was also an allegiance to the critical tradition that shaped the canon, and that it was precisely this which attracted, and still attracts, so much attention to McGann and Levinson's 1980s writings. In this respect a revealing comparison can be made between them and Clifford Siskin, one critic associated with 'New Historicism' who can accurately be accused of attempting to explicitly devalue the 'Romantic' canon in the 1980s.  Siskin's 1988 Historicity of Romantic Discourse was noteworthy for moving well beyond The Romantic Ideology and representing McGann himself as still steeped in 'Romantic' values. What McGann had failed to realise, Siskin claimed, is that critics and teachers of literature cannot simply separate themselves from the 'Romantic ideology' and expect the professional discipline of literary studies to otherwise continue unchanged. As he has more recently put it:
The reason ... that Romantic discourse so thoroughly penetrates the study of Literature is that Literature emerged in its presently narrowed—but thus deep and disciplinary—form during th[e 'Romantic'] period and thus in that discourse. 
Siskin's claims were large and potentially very destructive of the 'traditional' study of literature. Yet he drew off little of the polemical fire reserved for McGann, Levinson, and (a name that by the end of the '80s was always associated with theirs) Alan Liu. The weight of hostile criticism these scholars received (and indeed continue to receive) was therefore not proportionate to the extremity of the critical positions they assumed, for it cannot be the case that Siskin's arguments were less threatening to interpretative orthodoxies. Rather, the weight of that criticism points to how central the works of the 'New Historicists' immediately seemed to be in 'Romantic' studies. And this, as suggested above, was not just because of a continued commitment to the idea of the 'representative' consciousness—that commitment explains familiarity rather than centrality—but because of the clear affinities of their work with the most powerful and influential established criticism. As McGann, Levinson and Liu all wrote on Wordsworth, it is worth emphasising their profound debt to Geoffrey Hartman, for example. Levinson even concluded her study of Wordsworth, rather remarkably, with the words: 'I end where I began, with Hartman, who was, I think, righter about 'Peele Castle' and all of Wordsworth than he perhaps knew'.  Hartman's model of a Wordsworth torn between Nature and Imagination served them well, for if Imagination involved the sacrifice of Nature, it was not difficult to add that it involved the sacrifice of social life. This is not intended to demean their criticism, for they were all well aware of their debt to Hartman, but to question the value of many of the attacks their work encountered. Those attacks largely failed to note that they were dealing not with some hopelessly errant school of thought, but with canonical criticism in every sense of the word. That is to say, the 'New Historicists' were concerned both with canonical texts and their best canonical interpreters. Hence the power of their writing: had McGann, Levinson and Liu devoted their 1980s work to examining the political commitments of (say) Samuel Rogers' poetry their criticism would have had little of the power and ability to generate discussion it manifestly had when applied to Wordsworth's.
In a sense, then, the claim that McGann, Levinson and Liu were opposed to the canon is a mark of profoundly baffled hostility. Their work actually says much about how the canon empowers a strong criticism, which is in fact the pragmatic way of defining what the canon is—those works which empower strong criticism. This power is best measured by its resistance to the power of the strongest criticism which seeks to explain it (a resistance that then empowers a new strong criticism, as in the case of 1980s 'New Historicism'). In this connection McGann himself has recently pointed to the 'vitality' of great literary works which allows them 'to survive the critical abstractions we bring to elucidate' them.  Convincing cases for unjust neglect typically feed on this power too. In 1989 Butler made Thalaba seem interesting and important because of its relationship with canonical works that she could assume her readers knew and were interested in. Take away the support of 'relationship' and there is little left. The case for the 'unjust' neglect of Southey may be very convincing, and perhaps it would be a commendable thing to study his work, but the fact is that criticism is naturally 'aristocratical', as Hazlitt said of poetry, and not 'republican' (Howe, V, 347): it is attracted to established power.
Nevertheless, the 1990s saw a striking rise in the importance attached to 'relationships', which is evidence of the massive influence Butler has had on the critical practice of 'Romanticists'. It has been urged throughout this essay that the conservative reaction to 'New Historicism', emphasising difference, concealed real affinities while obscuring other differences that have since grown in importance. The difference illustrated between Levinson and Butler in 1989 has now become central: it no longer seems helpful to divide the field between, say, Abrams, Bloom, Hartman and McFarland on one hand, and McGann, Butler, Levinson and Liu on the other. Rather, it is necessary to accept a new 'key' division between the 'history of ideas' approach, requiring wide knowledge of post-Enlightenment philosophy and aesthetic theory (Levinson), and the historical approach (not, it has been argued here, to be confused with 'New Historicism'), requiring the widest possible reading in British culture (broadly understood) between the 1780s and 1820s (Butler). Although the tendency of the former approach since the 1960s to subsume modern 'Literary Theory' has led to theoretical modifications being mistaken for major procedural differences, in fact writing about Coleridge in the light of Spinoza, or Wordsworth in the light of Marx, is procedurally very similar: in both cases dialogue is constructed over national, chronologic, and generic boundaries. The real difference is with the sort of criticism that constructs a dialogue between, say, Gothic novelists because they inhabit the same national and chronologic frame.
The historical criticism of the 1990s represents a more profound threat to the 'Romantic' canon than the 'New Historicism' of the '80s, for reasons examined in regard to Butler's 1989 essay above. This threat is not likely to prove long-lasting, however, for the subject of 'Romanticism' cannot be indefinitely broadened without some very solid support at the centre (and this is a bad time for over-specialisation given that the period is increasingly being carved up at the institutional level). One notable difference between the '80s and the '90s, it is already clear, is that the former decade produced several books, like Butler's Romantics, Rebels and Reactionaries and McGann's Romantic Ideology, which generated a huge amount of discussion, while the latter decade, by comparison, did not. This points to the problem of audience, and the possible loss of a sense of communal participation in an important critical and scholarly venture. The reason, of course, is that historical criticism of the Butler stamp has no obvious focus except, rather vaguely, a historical period. In firmly rejecting the idea of the 'representative' consciousness, it often seems to assume the loss of the canon as a consequence and then concerns itself with tracing out increasingly fugitive patterns of cultural correspondence. But the connection made between (supposedly obsolete) 'representative' consciousness ways of thinking and an artificially restricted canon is a glib one. In her 1989 essay, for example, Butler writes:
The recent American Romanticist orthodoxy declares the great Romantic topic to be the alienated individual consciousness; the great work, Wordsworth's Prelude, that autobiography of a post-revolutionary recluse. ... But the heroic ideal being extracted from Romantic poetry—the way of the literate recluse— is far too privileged and—dare one suggest—too professionally interested, to seem truly universal. It needs not ousting but supplementing, with forms of poetry and novels that are serious and intelligent without so often being private and academic.81-2
It is worth recalling Macaulay's famous response to The Prelude—'The poem is to the last degree Jacobinical, indeed Socialist'—just to get some purchase on Butler's smooth equation of The Prelude itself with a particular critical 'orthodoxy'. Her remark makes it clear why 'the canon' nearly always gets a bad press in this sort of criticism: if the champions of 'the canon' are conservatives, 'the canon' must be conservative too. A (possible) problem at the critical level, it is alleged, must be answered at the institutional level (courses must open themselves to new 'forms of poetry and novels'). The proper question, perhaps, is not whether the canon should be extended—it has been, and no doubt will be—but whether this is a good reason for extending it. The canon is not, after all, a smug, plushly-bound, clearly-defined collection of 'safe' classics. To defend or attack it is to fetishize it, to impose a settled shape and substance on what in practice has neither. Like other anti-canonists, Butler tends to make the canon 'Wordsworth and Coleridge' when it suits her. But in fact the 'Romantic' canon (preserving for the time being this misleading designation) was, and remains, singularly open, involving many unanswered questions, just the first of which is how the balance of power between Blake, Wordsworth, Byron, Scott and Hazlitt should be adjusted. Ignoring this unstable energy and pretending that the canon really is no more than its most conservative defenders would have it be makes for an artificially 'liberated', 'republican' criticism. The challenge ought now to be to recognise that the canon and the 'spirit of the age' are separate concepts, created for separate purposes, even though 'Romantic' period studies have often seemed determined to align them (the problematic case of Jane Austen notwithstanding). That is to say, the claim for Wordsworth being the most powerful poet of his generation is quite distinct from the claim that his was a historical period's 'representative' consciousness.
I want to end here by speculatively setting McGann against some of today's historical critics. His is a thought-provoking case, for he seems to have moved through two distinct phases in the last few years. The first is represented by his controversial New Oxford Book of Romantic Period Verse (1993) which aimed, like Butler's 1989 essay, to 'throw many of the old familiar poems into new and interesting relationships', while actually throwing many of the 'old familiar poems' away.  A little curiously, for the author of The Romantic(Ideology, he suggested in the introduction that 'an anthology of writing from the romantic period may be a better resource for study (and even for understanding) than a more theoretical work' (McGann 1993, xix). This was McGann at his closest to Butler and the historical school. But the second phase, represented by The Poetics of Sensibility (1996), indicates a substantial qualification of this position. Here McGann protests at untheorised recovery operations: 'such work ... [has] a tendency to evade the question of the aesthetic character and value of the obscured texts; or, if those questions are addressed, to look for value in the moral qualities that can be found in the works' (McGann 1996, 4-5). This is a fascinating statement, not least because it sounds so much like the charges leveled against McGann and the 'New Historicists' in the 1980s by critics like McFarland. One senses a movement that is forward, but also circling. In the course of the same discussion McGann notes that he is not concerned with 'change or stasis in the canon of what we read', but again, as in The Romantic Ideology, with how we read that canon (McGann 1996, 5). These remarks are the more germane in that they do not presuppose a fixed, consolidated canon, and they actually encourage an alertness to the critical possibilities latent in work not in the canon. But the question 'how?', even when modified by the question 'what?', should not be subordinate to it, and a theory of reading in its naturally dialogic aspect requires application to a body of writing to which other theories have been applied.
Prompted by the 50th anniversary of the first publication of Foucault’s famous book commonly known in English as Madness and Civilization, this essay explores how the book has changed between versions, in the process losing what can be cast as both its phenomenological undertones and a ‘romanticism’ about the truths supposedly revealed by madness. Reasons for Foucault’s own disavowal of these elements are considered, and taken together – conjoining a critical biography of the book with attention to Foucault’s reactions to it – this essay fashions a mirror to hold up to certain currents within contemporary human geography. It is argued that the ‘romantic fantasy’ which permeates the original book, if not overwhelming it, has significant echoes in the ‘romantic gesture’ displayed by some present-day geographers. The older Foucault’s distancing from his earlier romanticism is hence instructive for scholars critiquing the recent history of human geography, but there may also be grounds for claiming that it would be mistaken to lose this romanticism, together with its phenomenological correlates, entirely.