RE-READING NORTHROP FRYE: IMAGES OF CULTURE AND THE CANADIAN CONTEXT

Responding to a prevailing critique of Northrop Frye's studies of Canadian culture, the author of the paper combines Frye's Canadian essays with his general theory in order to demonstrate that Frye's use of the term imagination, similar! y to Iser' s notion of "imaginary", allows a differentiation of the social imaginary and the fictive. Frye has provided a background for treatment of the fictive as aesthetic structure and has employed the study of Canadian culture primarily as a specific tool to describe literature with a disturbed aesthetic structure .. INTRODUCTION & THE CONTEXT In this paper I discuss Northrop Frye's theories in the light of the thorough going critique of mythopoetical method after the rise of Canadian anti-thematic criticism. In my opinion, these attacks were hrrgely an attempt to earn symbolic capital, and consequently, the dominant position in the Canadian literary market.1 During this process Frye's theory was too often reduced and simplified. But his practice has to be treated in a broader context. It may, for example, be interpreted within the newer notions of the imaginary and the social imaginary. This interpretation can also provide a link to more formal aspects of literary studies. Through this kind of interpretation I shall try to demonstrate that critique of Frye's method has not always been justified. Although it was certainly aware of it, Canadian criticism somewhat neglected Frye's general theory, despite its international acclaim. Critics in Canada noted the success of the Anatomy of Criticism or of The Great Code; but there were few at1 Literary practices and practices of criticism are inevitably part of a "Bourdieauan" literary market which tends to homogenize the literary community and its communications. They may be interpreted as orthodox and heterodox utterances in the field of contesting critical identifications and interpretations whose value depends on their symbolic capital, as well as on the existent relations of power (Bourdieu 66-67, 71-76). Formalist and poststructuralist critique of Frye's theories signified a change in the Canadian literary market. This view is supported, for example, by Lecker's reading of Frank Davey's Surviving the Paraphrase. Davey's critique is shown as an unorthodox utterance which started the battle to overpower the dominating position of the thematic criticism. (Lecker, sec. Ill, sec. IV).


INTRODUCTION & THE CONTEXT
In this paper I discuss Northrop Frye's theories in the light of the thorough going critique of mythopoetical method after the rise of Canadian anti-thematic criticism.In my opinion, these attacks were hrrgely an attempt to earn symbolic capital, and consequently, the dominant position in the Canadian literary market. 1During this process Frye's theory was too often reduced and simplified.But his practice has to be treated in a broader context.It may, for example, be interpreted within the newer notions of the imaginary and the social imaginary.This interpretation can also provide a link to more formal aspects of literary studies.Through this kind of interpretation I shall try to demonstrate that critique of Frye's method has not always been justified.
Although it was certainly aware of it, Canadian criticism somewhat neglected Frye's general theory, despite its international acclaim.Critics in Canada noted the success of the Anatomy of Criticism or of The Great Code; but there were few at-tempts to include it in the discussion of Frye's criticism of Canadian literature (Rajan 134; Cameron 114). 2 Frye's position was largely defined domestically by his "Canadian" essays.The literary field was dominated by his Bush Garden, by the Conclusion to the 1965 project of the Literary History, by his essay "On Canadian Poetry", and by The Educated Imagination.Indeed, one significant reason for the neglect of his general theory lay in a failure to distinguish his practice and the closely related thematic criticism.
There was more than one reason for the amalgation of these two aspects of mythosymbolic approach in Canada, but the principal reason was the dominant anti-American cultural nationalism of the 1960s (Brown, Practice, sec.II).Certainly, the problem of Canadian national identity has served to narrow significantly the understanding of Frye's theories, since literary production and the research of literature programmatically emphasized the problem of national identification. 3Thematic studies have been clearly indebted to Frye's concepts of garrison mentality, 'evocation of stark terror' and, most of all, to the notions of a central unity in literature and its underlying national imagination.This is particularly evident in the early works of M. Atwood (Survival), J. Moss (Patterns of Isolation in English Canadian Fiction) and D.G. Jones (Butterfly on Rock).However, Frye's approach cannot be interpreted in the same manner as their efforts.The most obvious characteristic of thematic criticism in its relation to Frye is the definitiveness by which Frye's notions were adopted.Even though Frye wrote merely of the "provisionally called" garrison mentality (Conclusion 830) and although he only occasionally used the term "theme", both terms have become central notions of thematic criticism.
Even if the research was not necessarily thematic (Cameron 114), the uniqueness of Canadian identity became one of the leading aspects in literary research in the years 1972-84.Already in their early article (1976), Cameron and Dixon maintain that it is a debatable point as to whether Frye's remarks are responsible for the "critical anachronism of thematic criticism", and argue that some of his followers ignored "the liberal spirit" cifFrye's general theory.It is in this context that we must read Cameron's later remark on the minimal influence of Anatomy of Criticism in the criticism of Canadian literature.
. The first attacks aimed at nationally oriented criticism followed rather rapidly, in the years 1976-1978.But with a broadening of the methodological field, especially with the rise of poststructuralist approaches, it became clear that the temporary dominance of thematics in the 1970s had caused an inability to confront both thematic criticism more broadly and, particularly, Frye 's general horizon.Russel MortoU: Brown thus reasonably draws attention to Laurie Ricou's warning in 1978 against establishing an anti-thematic orthodoxy "from which no variation is allowed" (Brown, Practice, sec.II).There were few attempts to surpass this orthodoxy.Brown's own essays are one such attempt, as well as the articles of Laurie Ricou and, in the late 1980s, the writings of T.D. MacLulich (ibid.).Marginalisation of thematic criticism provoked attacks on Frye's "Canadian essays". 4Elsewhere, it provoked a considerable ignorance of his broader mythosymbolic approach.In my view, neither of these developments was justified.5While the critique of thematic criticism, as well as of Frye' s methodology, was far from being uniform, two distinct charges were arising: 1.) At the outset, thematic criticism was attacked by Frank Davey in 1976, and by Cameron and Dixon in 1977.Their main accusation was that such criticism focuses on the "social and historical setting", rather than on literature.Davey and Cameron each repeated their accusations in the 1980's and 1990's, and were followed by W.J. Keith and several others (Brown, Practice, sec.II).Since most of the thematic studies have actually dealt with the specific cultural material present in the literature, it is correct to say that they have implicitly become cultural studies .However, cultural and sociological aspects of literary research have long been recognized as legitimate methods in the pluralism of literary research (Virk,sec.l,22),and have been thoroughly popularized with the rise of Empirical Studies of Literature and with the growing influence of cultural studies.Frye's general theory, moreover, has never reduced litetatl.lteto its cultural aspects without considering its structural dimensions.As I shall demonstrate, in his work these two aspects are intertwined in a complex and thoroughly modern way.
2.) With the rise of deconstructionist criticism, the urge to define a single fixed cultural identity was also attacked.Mostly it was discarded in favour of "un-fixed" notions of regionalism and ethnic literatures, as is evident, for example, in the works of R. Lecker, Alison Conway, Roy Miki and E. Mandel (Brown, Practice, sec.II).These views correspond to so called anti-essentialist tendencies in cultural studies which reject any fixed identity, including any environmentally based identity.Since environment ought to be regarded as a part of "un-fixed" social process and its discourses, one ofthe specific postructuralist charges addressed to thematic criticism was the charge of environmental determinism-that is, of treating "the landscape as given" (Cameron 114).
Even though both charges might be applied to Atwood's, Jones's and in part to Moss's criticism, they may not be as easily directed to Frye.Frye's criticism cannot be thought outside his general theory which is not limited to dealing with culturally specific material.Frye focuses on the general pattern in the genesis of a specific literature, comparing its structure to the structure of myth, which is not necessarily the 'foundational myth' of a culture or of a specific country.This proto-structuralist orientation was acknowledged quite early (Brown,Critic,180).Along with it, however, Frye's criticism is also grounded in the analysis of image.Imagination is viewed as the reflection of cultural values.Frye's main interest, accordingly, is the analysis of common symbolic structure of the imaginary.This does not, however, imply any fixed identity.The notion of imagination as a symbolic unity inevitably permeates his understanding of the Canadian imaginary and I shall demonstrate the complexity of this thought while focusing on the relation of art and criticism to society.THE IMAGINARY Although Frye's remark that the role of a critic should be "to translate literature into a continuous dialogue with society" would seem to confirm the notion that "all criticism is social criticism" (Myth 138), Frye treats cautiously any literary activity, including criticism, that seeks a direct transformation of its ideas into society or, conversely, of social 'voice' of ideas into the field of imagination.It is evident that he treats art as a constitutional element of a homogenized society, since it is his view that literature should "inform and reform the society'' (Myth 305-06), but it is also evident that art does not need to be culturally ideological, for "in a fully mature literary tradition the writer enters into a structure of traditional stories and images", and the speaker becomes "a place where a verbal structure is taking its own shape" ( Conclusion 835).I find these remarks very important, because they are closely related to Frye's system of metaphorical and descriptive language.While descriptive language is focused on reference outside its own structure (Great 46, 61-62), imagination is regarded precisely as the ability to enter into a metaphorical structure itself.It is regarded as something related not merely to literature or to the arts, but to the primary creativity of an individual as well as of society.In this context, literature, like mythology, does not exist to describe specific conditions, but to contain them in a manner that does not limit significations in the structure (ibid.; cf. also Conclusion 836).This, as we shall see, is the concept developed in more recent theoretical notions of the imaginary.
As paradoxical as it may seem in view of antiessentialist attacks, Frye's ideas most likely result from the conviction that society is a construct, a 'collection' of mental representations, rather like any identity.Cameron's claim that thematic criticism treats landscape as a given rather than as a social construction (cf.also Pontuale 41) is therefore only partly justified.Already in the famous Conclusion, Frye implicitly acknowledges that the foundational myth is essentially a social-subjective construction.It is true, however, that social construction stems from the existing historic or archaic structural stratum, which is understood as a verbal or imaginative structure.Frye, therefore, consistently regards society as a set of mental images based in established social models and patterns, and in so-called stock responses (Stubborn 7, 20).The stock responses, particularly, are shared through mass media, interpersonal con-versation, textbooks and, possibly, through literature and literary criticism.This notion also permeates his treatment of the Canadian foundational myth.Being an instrument of mental production, art, like any other human experience, is caught in the relation between the society as a whole and the variety of its individual mental forms.It is a relation which Frye openly associates with the primitive function of religio: "of binding together a society with the acts and beliefs of a common concern" .
At this point, in particular, Frye's thought resembles the ideas of European theorists such as Cornelius Castoriadis or Wolfgang Iser.The resemblance is most obvious in the binding of the construction of society and of an individual to a previously existing structure.In all of these conceptions, the imaginary and the social imaginary become to some extent 'religious' bases; they become unifying principles which bind together individual representations.At the same time, art, science, scholarship, mythology and religion, as well as any other human experience, become parts of an integral society based on the imaginary.Castoriadis, in particular, describes the imaginary as a creative principle, which manifests itself in the society and defines its organization.In his view, any social act and any social existence is thought, described, and achieves its meaning only through a net of significations.The chief function in society is given to what Castoriadis calls central imaginary significations, such as God or nation.The central imaginary significations do not refer to any real or rational signified; instead, their 'referent' is produced precisely through the institution of the central imaginary signification and is at the same time manifested in individual representations of the signification.Thus, the central imaginary significations at once create objects of individual representations and become a regulator of possible imaginary meanings circulating within society (Castoriadis 140,362).
Castoriadis's idea is interesting above all because he understands that the social imaginary may become manifest only through symbolic means.Herein lies a chance to understand another problem which is at least partly associated with Frye's theory.Even his general theory cannot be thought without considering the problematic relation between the social imaginary and ideology.It is clear that ideology is closely related to the fact that each society inevitably expands its own signifying structure with its own symbolic means.Thus, Claude Lefort was able to develop the idea of the social imaginary using the classic distinction between traditional and modern society, and arguing that the manifestation of the social imaginary in modern Western society is dominated by ideology and its structures of power (Lefort 281,. It is nevertheless too limiting to confine ideology solely to capitalist society as Lefort does.Castoriadis himself understands that we are able to comprehend the social imaginary exclusively through the notions of our own culture.We are restricted to the framework of our own organising system, which is dominated by the 'logique ensembliste-identitaire': a logic which distinguishes, determines and posits in order to assemble, adjust and construct.In as far as it is based on the authority of logicalrational argumentation, this 'united and distinct' logic is in itself a unifying principle .But it has to be added that real domination of power is always also based in the institutional position of an argument.It is the institutional position of an utterance that enables a forced unification, since it enables, initiates and preserves the mobilisation of signification by use of linguistic or symbolic logic .Ideology is therefore based on rational-argumentative logic, which corresponds to Castoriadis's 'logique ensembliste-identitaire'.However, its unifying agency is enabled only through a traditional metaphysical view of the 'domination of the world', and a suppression of ambivalence which is a characteristic feature of the modem society (see Bauman; Debeljak).
The central question in this respect considers the state of literature and arts in the conceived system of the 'domination of the world' -to be more specific, a question as to the strength of their potential to become vehicles of homogenisation and ideological domination.It seems that the best possible way to begin to address this question is to understand the potential of the fictive component in the arts, since art is also rooted in the imaginary itself.
Castoriadis and Frye differ considerably on this point.While Castoriadis envisages the possibility of an imaginary which is not necessarily manifest through rational logic, Frye still perceives organisation and domination as the only possible human relation to the world.In the Anatomy he apparently presupposes the inability of an individual to grasp criticism as his object in its totality.In his conception Frye proceeds from the traditional view that the object is something which is, in advance, summoned to the individual's autonomous systematisation and comprehension.It seems that in this perspective even the 'verbal structure', which is a common centre of myth and-literary shaping, is understood as something metaphysical (Concluston 835-37).Therefore it is not surprising to come across the opinion that the arts, albeit by means of a creative dialogue, are also always subject to organising (Stubborn 16-17).Such a stance implies that all the instruments of mental production, including the arts, may become vehicles of ideological unification.Frye nevertheless distinguishes the arts from other bodies of knowledge (Stubborn 20), and this seems to be an important difference.Contrary to science, arts and scholarship should not only be able to grasp the total truth of the objects they lean on, but are to be left to shape themselves (Stubborn 51, 20).Through instituting form, the arts, instead of recreating it, participate in the shaping of reality as it is perceived.In considering some of Frye's remarks, we might expect to find the '1-Thou' dialogue to be a basic principle of this participation.
The dialogical principle in the arts is in fact crucial in determining their role, since it implies a distinctive manifestation of the imaginary.Wolf gang Iser has shown that the imaginary is a potential activated from outside itself, which discloses itself differently according to its manifold activators, but which always unfolds itself as play (lser 222-23).Social constructs, religion, individual imagination, reveries, and, finally, the fictive, are therefore merely its manifold manifestations.The fictive component -which is, of course, at the core of any art -is thus seen as one of the activators of the imaginary.Certainly, its orientation is significantly different in comparison with the social imaginary.The fictive component has "far less of pragmatic orientation", and since it is "a means of making the imaginary accessible to experience outside its pragmatic function", the fictive establishes a specific relation to material reality.In Iser's view, the particularity of this relation is precisely its dialogical character: the fictive is manifested as a 'doubling structure' .
Frye already recognises the duality of literary texts, a duality that is attributable to all the arts, although it is, in his view, primarily a linguistic duality.He treats descriptive language as principally centrifugal, and while descriptive language is primarily oriented to the reference outside the verbal structure, metaphorical language is above all centripetal.It is focused on the shaping of the structure itself, and it is particularly focused on the structural interaction of its constitutive elements (Anatomy 74-75, Great 46, 61-62).Proceeding from Frye's theory of modes and adjusting it to his theory of fictional narrative, Paul Riceour observes similar dualities, embodied primarily in a bipolar interweaving of the narrator's time/discourse and the time/ discourse of literary personae (148)(149).Both types of discourse are, however, enmeshed in a complex relation.The textual world, as formally closed as it may seem, is always also open to the world outside itself.It is constructed in the intersection of the textual world and the world of ordinary action (Time 100-101).
Frye first anticipated this textual duality in the Anatomy of Criticism.In the theory of modes he articulated the notions of 'internal' and 'external' fiction which correspond to the fictional and the thematic modes of literature.Perhaps it is too limiting to understand 'external' fiction exclusively in relation to thematic material and to the theme of a literary work.Nevertheless, it is obvious that by distinguishing heroes and their societies from the communication model of the author and reader, Frye implied a fundamental difference between the textual and the extratextual world (Anatomy 52).It is a difference that matches the idea of a centripetal and 'a centrifugal force at the level of the metaphor.Frye' s own theory of symbols proves how both tendencies must be present in each literary structure, although they are related through a specific tension (Anatomy 73-75).Yet Ricoeur demonstrates convincingly that in this context the literary symbol becomes a 'hypothetical verbal structure' wherein the centripetal force prevails : this distinguishes fiction from other narrative modes and, in fact, activates what lser has called a 'doubling structure'.
The particular distinction of fictional narrative, in view of its dialogic character, therefore, lies in the specific relation between the textual and the extratextual world.It is important to note that when 'the world outside' is 'transformed' into fiction, it does not exist in its primary mode of being.In fiction, neither the textual nor the extratextual world is significant in itself: both worlds -as signifiers -are drawn in mutual readings through each other's eyes.The reader's attention is thus drawn to signs as signs, which cancels any preestablished correlation between signs.In short, they no longer refer to anything given, but become at best the signified of the other signifier.They become interwoven into circumreferentiality, activating the imaginary by inscribing the two coexisting discourses one in another.In this way a structure is enabled where "every word becomes dialogic, and every semantic field is doubled by another" .
Despite this specific concentration on the activation ofthe 'play' which puts the reading of both worlds under the sign of 'as if', the structure of fiction is never absolutely innovative.Frye properly observes that it is always to some degree determined by the conventions of literary shaping.In his own words, literature is dependent on the set of "huge containing conceptions which establish the literary societies".This means that it is preconditioned by the verbal structures from which it derives and which Frye identifies as myths or, rather, mythoi (Stubborn 53; cf. also Anatomy 162).Not surprisingly, since they have each taken it over from Aristotle's Poetics, his conception of mythos is very similar to Ricoeur's latter notion of narrative: mythos is in both cases understood as a discursive organisation of narrative events.Following Ricoeur's thought, however, it is worth noting that in western society mythos may be manifested either as historical or as fictional narrative.I have shown that in the latter case its composition is a composition of a 'doubling structure', which means that any preconditioned structure is in itself subjected to the fictional play of 'decomposition' and 'composition'.In the historical narrative, on the other hand, the centrifugal force bases the narrative in the pragmatic orientation, validating its meaning almost exclusively against the grain of extratextual reality.Thus, the imaginary is activated in another type of play: as the social imaginary.As I have also argued, the social imaginary in modern western society is predominantly activated in the play of hierarchical positions; it is manifested in a structure of discourses subjected to the relations of power.Hence, its homogenising potential is usually ideological. 6Having the same structural precondition, as Frye demonstrates, literature as a verbal structure may conversely exert an influence on the structure of mythos itself, either by changing it or by actively preserving the existing relations.
In the Conclusion, Frye describes the major part of pre-Centennial Canadian literature as one which validates its meaning against the grain of extratextual reality.He defines it as a literature which "heavily stressed the conceptual and argumentative use of language", and whose writers did not "naturally think metaphorically but descriptively" (836).We might argue that this kind of literature may have exerted an influence on the homogenising structure of the Canadian social imaginary, and that in the final consequence it may have become an element in the ideological relations of power.The possibility for such a 'passage' ofthe fictive to the social imaginary opens up at the point where literature, instead of activating a fictive dialogic play, directs the reader to seek identification within the signifiers of the 'real', extratextual world.Again in the Conclusion, Frye describes this type of literature as literature that encourages the reader "to remain within his habitual social responses" instead of prodding him into the autonomous imaginative world (838).Although in the paraphrased passage Frye speaks of Canadian "popular literature", this observation applies to any literature which refers to the extratextual world and destabilizes the doubling structure.I should add here, that when the centrifugal force in the structure becomes dominant, the fictive narrative turns into historical narrative and to a method of organising the meanings of a culture.Canadian fiction, therefore, may become a tool to confirm a specific cultural identity.More to the point, Frye evidently understands that this is not merely a typical feature of popular literature, but of any literature which destabilizes or fails to create the doubling structure.This is why he declares that literature should not describe specific (extratextual) details, singular themes and mythical stories, but should contain them in the adopted metaphorical structure.It should become "a place where a verbal structure is taking its own shape" (e.g.Conclusion 836).
On the one hand, of course, this is the point where he finds Canadian literature lacking.On the other hand, this is the point where he finds it most representative.He is aware that the imaginary inevitably manifests itself through 'verbal' structure, and that it is expressed through symbolic and narrative conventions.Through the Canadian case Frye actually tries to demonstrate the interconnections between the verbal structure as literary convention, the verbal structure as shaping principle of an individual literary work, and the verbal structure as the organizing principle of society.In the Conclusion, the critique of Canadian literature (if this could be considered as critique) originates from a clear understanding that literature should produce its own metaphorical, doubling structure, instead of belonging to a correspondent mythology produced by the society (842).

STRUCTURES AND CONVENTIONS
Frye's complex thought on imagination at one and the same time moved him into proto-structuralism and anticipated the kind of socio-anthropological thinking which-was later clearly exemplified by Castoriadis and Iser.Since this thought was already present in the Conclusion, we may say that it infused his criticism of Canadian literature from the very start.If Frye's interest was thus not focused on the thematic material of a fully mature literature, the genesis of a foundational myth -the 'incompleteness' of Canadian literature-was for him a possible field in which to explore a social mythological structure.This means, in fact, that his interest in Canadian literature was precisely an interest in literature with a disturbed aesthetic structure (e.g.ibid.821).
Of course, his thought implies neither that Canadian literature would remain in an incomplete phase, nor that the analysis of structure should remain concerned with this initial phase.In the critique of Frye's criticism this "detail" has often been overlooked, although it is a basic presumption of the Conclusion."The conception of what is literary has to be greatly broadened for such literature," warns Frye, and thus "it is more significantly studied as a part of Canadian life than as a part of an autonomous world of literature" (822).It is obvious that such a conception of literary study is a useful tool only to explain why literature within a certain social context has not been able to produce 'mature' and non-imitative aesthetic forms.At best, such a conception may help us understand why literature has been used as an ideological apparatus -as one of the elements of power to uphold a unified cultural identity -instead of as a fictive manifestation of the imaginary.
ldo not mean to suggest that Frye was interested in researching the ideological function of Canadian literature.By understanding this presumption, however, many of Frye's apparent inconsistencies become less contradictory.Failing to appreciate this is, I think, one of the reasons notions such as that of the 'Canadian North', the 'evocation of stark terror' or the 'garrison mentality', have so often been disputed.
Their rebuttal as critical tools implies at least two problematical stances.The first of these is one that does not allow for cultural criticism as a heuristic tool, albeit limited, which applies specifically to analysis of literature as social mythology.By disallowing this, this stance dismisses an instrument for.the analysis of literatures that remain predominantly in the domain of the social imaginary instead of being able to activate the fictive.In fact, this stance implies a doubt as to whether there has been any literature in Canada that was predominantly in the domain of social imaginary.The second of these problematical stances is one which fails to take into consideration Frye's own epistemological distance in regard to these notions as conventions: as sets of stock responses which primarily function as homogenisation tools, even as ideological impulses.This stance may be readily demonstrated in regard to the charge, often addressed to Frye's criticism, of environmental determinism.
In his early article, R.M. Brown has shown that even Atwood's "victim positions" were a kind of grammar, "important in that it does not direct our attention to { ... } the raw material of environment { ... } but to the structures given that experience" (Critic 180).The fact that nature, "wilderness" or "northern frontier" were given as conventions cannot be ignored.They were attributed to the opposition between the reality of the encountered land and inherited European conceptions, or even between the new land and the old language (Briscoe- Thompson; cf. McGregor etc.).It would, therefore, be wrong to conceive of this opposition as materially determined, in the sense of 'environmental determinism'; since it was-manifest from very early on as convention.It was rooted in many different stock responses, including the loyalist imaginative response of trauma and injustice  and the puritan identification with the Israelite's exodus .
The notion of "wilderness" as convention is present as early as in Frye's Conclusion.In this famous essay he warns that "the mystique of Canadianism" was so full of wilderness because the Confederation came too suddenly after the pioneer period to create further stock responses for cultural identification (826).Frye, of course, traces other conventions that shaped Canadian imagination and literature: e.g, the view of Indians, who "like the rest of the country, were seen as nineteenth-century literary conventions" (836).His comments on cultural colonialism, with its two sources of England and USA respectively, indicate a further pole in the register of stock responses.Along with constant traces of imperialist logic, one of the main sources of these patterns and responses was clearly, as C.D. Mazoff has shown, the logic of economic progressivism and mercantilism.
Nevertheless, one of the most interesting critiques of Frye's presumed determinism was Lean Surette's charge in the essay on topocentrism two decades ago.Surette argued that Canadian literature, if it had been rooted in different topography alone, would have been severed from European culture and tradition, unable to invent new forms or to renew in a fruitful way the old 'European' aesthetic models.But I find his assertion simplified, since it does not take into account the fact that topography, as part of the material stratum, is only one of the layers on which the imaginary leans.The narrative structure-or rather, the narrative precondition of the cultural myth-is indeed established upon this layer.Literature as a fictive manifestation of the imaginary, even though it is rooted basically in the same structural precondition as myth, may, however, take the cultural myth as its source, or as its target, rather than as its condition.In both cases, the imaginary manifests itself only through convention, or as breaking of convention.Therefore it is not necessary to place these two discourses in opposition, as it is hardly necessary to place in opposition the 'new' culture and the 'old' tradition, in particular, the new topography and the aesthetic models or the conventions taken from the "old" European tradition.
Frye clearly states that by form he does not mean external form, but an internal shaping principle, i.e. the wholeness of a structure (Bush 175-76).Especially in literature, formal models and conventions may therefore be taken over from elsewhere; however, they may only be transformed by adopting the same inherent structure of the mythos into which they enter.As I have shown, in fictive discourse this 'structure' is further reshuffled and reshaped in a play of decomposition and (re)composition.The refusal of the possibility that the European or British traditions would inform the verbal structure of the 'Canadian myth' could thus merely indicate that the British tradition was not able to function as a general organizing principle of the society -as the central imaginary significance, to use Castoriadis's phrasing.It does not mean, however, that singular formal models and conventions in culture or in fictive discourse would be unable to function in different structural preconditions that inform a different specific literature.
It is in this context that we must read, for example, Frye's remark on the protoform and the proto-sources ofCanadian literature (Conclusion 835; Cf.Hop wood 20).Inscribed in these proto-sources is the superior-inferior conflict between European and indigenous traditions, which is often believed to be the main area sustaining a verbal structure ofthe Canadian cultural myth (New 18-23).Considering this inevitable presence of the European tradition, it is in fact impossible to expect that Canadian literature would not adopt European aesthetic models.Nevertheless, it would be equally impossible to have expected these models to be used in a non-imitative way, unless they were adopted into its own 'verbal structure'.As long as Canadian literature accepted these forms without adapting them to its own preconditioned structure (whether this structure existed or not), it was bound to use them not only in an imitative way, but also as a means of sheer thematic description.Literary endeavours were limited to reference to extratextual reality, for without 'entering' into the structure of traditional stories and images, as Frye terms it, a fictional work is not able to use it as a precondition to be activated in an interplay between the fictive and the imaginary.
If we accept these notions, it is easy to see that records and journals of the first explorers may be the proto-source of Canadian literature only as a set or a register of models, patterns and stock responses, rather than as a direct source of any literary structure.This also applies to the oral tradition of the indigenous people, the Jesuit Relations, or to the texts of the sisters Parr Trail.Canadian literature and Canadian imagination could be structured only as sets of conventions which were construed and re-construed, interpreted and re-interpreted, which became intertwined etc.As constructions they were far from being unchangeable: they are, on the contrary, unstable identities.Even for the garrison mentality Frye claims that... "as the centre of Canadian life moves from the fortress to the metropolis, the garrison mentality changes corrrespondingly" (Conclusion 834).Frye was obviously aware of the mutability of the imagination which was later explicited in Castoriadis's notion of the imaginary.For Castoriadis, the imaginary is always open to change; it is subject to transformation precisely because it also creates its own eidos.Frye's "imagination" might be more rigid, but it is still not fixed or in any way stable.

CONCLUSION
It is therefore clear that Frye's interest in Canadian imagination is rooted in his belief that Canadian literature was for a long time predominantly unable to use conventions as mere preconditions for the activation of fictive play.Frye's structural interest was in researching the resemblances between the typical structure of Canadian literature and the structure of Canadian social mythology.It was primarily an interest in similar manifestations of the imaginary.Indeed, in his Conclusion Frye argues that there exists "such a thing as an imaginative continuum", and that "writers are conditioned in their attitudes by their predecessors, or by the cultural climate of their predecessors" (849).However, by doing so he obviously does not argue for a fixed, cemented and uniform cultural unity.He is aware that as a set of conventions and codes even this particular social-historical manifestation of the imaginary is open to changes and to indeterminancy.ltis,in_otherwords, a "culturalgen~ralisation", and as such it is only one of the heuristic tools.
Despite the fact that, as a heuristic tool, it is only temporizing fiction (Brown, Practice, sec.Ill), this is not enough to dismiss it.Every language is the abuse of language, argues Castoriadis; still, it is needed exactly because it is univocal sufficiently as to usage.Although identifications may not be fixed notions, we need unifying concepts to be able to live in a meaningful world of significations.As I have tried to demonstrate, Frye was aware of both tendencies and it is through this awareness that we must read his criticism of Canadian literature, as well as of Canadian society.