THESE REMARKS ABOUT THE CONCEPT of region are intended to be neither polemical nor definitive. They pose questions to which I do not pretend to have ready answers. They are inconclusive. All that I am convinced of is that there should be new places to talk about, and new discussions focusing on, such questions. And I should start off from being up-front about the Gramscian assumptions that have shaped my outlook, my own ambiguous identity as a scholar (but not a resident) of this region, and my own involvement in some of the positions on "region" I intend to probe.

The preliminary programme of this conference suggests the tremendous vitality and diversity of scholarship in Atlantic Canada.[1] This is a conference that reflects its day. A rough tabulation, based on the preliminary programme, suggests that of the 42 papers destined for this conference, no fewer than 15 directly reflect the emergence of gender as a category of analysis and of women as important subjects in history. A further six probe minority identities, those of race and ethnicity (four) and those of religion (two). A further ten look at economic history, most with a particular focus on local communities; and a further eight might loosely be described as studies in local state-formation and official culture. One cannot judge in advance how any of these themes will be developed at our workshop, but it seems possible to say that this is emphatically a 1990s agenda, placing emphasis on gender, ethnicity and community, rather than on, say, such 1970s and early-1980s topics as regional underdevelopment, class conflict and the position of the Atlantic Provinces within Confederation. We will be hearing a lot about Nova Scotia (19 papers), something about Newfoundland and Labrador (seven), New Brunswick (four) and Prince Edward Island (two); only four papers locate themselves within "Atlantic Canada", and only one within the "Maritimes". This is a pluralistic, perhaps even diffuse, set of topics; the old narratives of nation-building, the coalesence of regional grievance, the development of underdevelopment and the making of working-class consciousness all recede, and new stories of identities and ideals are now on the agenda. There is evidence here of a certain skepticism about the models, master narratives, grand theories and universal assumptions that appeared to structure earlier work, in this region as elsewhere.

Contrary perhaps to your expectations, and contrary to the positions taken by numerous of my labour-history and neo-Marxist friends and relations, I am not here to argue that either this topical heterogeneity or this skepticism towards the older narrative frameworks are mistakes. Rather, I see them both as "moments" in the development of regional historiography, as new ways of interrogating a past that will always, this side of eternity, elude a decisive and final theorization. In many ways, this opening up of the infinite diversity of the past, these long-suppressed voices, is a continuation of, not a break with, the social-history revolution of the 1960s and 1970s, and, perhaps at future Atlantic Canada Workshops, other marginalized voices — of gays and lesbians, of prisoners, of the First Nations, for example — might be more strongly present than they are here. It is not difficult for a Gramscian to applaud the politics of inclusion which has indirectly led to this new pattern. This ACW programme signals, perhaps unintentionally and perhaps not, a revolution in the paradigms through which we have made history out of the past. This moment seems, in many ways, a good one.

At the same time, this multiplication of subject-positions, epistemologies and methodologies places a question mark over the very concept of an "Atlantic Canada" whose construction we might meaningfully undertake at an Atlantic Canada Workshop. The sense that the past can be explored in completely new ways, using such categories (among others) as gender, ethnicity, race, age, sexual orientation and the environment, while liberating "history" from what had come to seem its outdated unilinear narratives of "development", seems to deliver us into a no less paralyzing experience of infinity. Every event in the past can be made part of a different interpretation, and these patterns of interpretation, each claiming to represent the whole, can leave us with a series of hermetically sealed parallel universes. Academic peace is achieved, after a fashion, by careful boundary maintenance. But the price tag is a diminished sense of the ways in which categories interpenetrate and illuminate each other. And less obviously, there is a greatly intensified tendency to turn particular categories of understanding into "forces" and "things" in their own right, in works which, because they are constructed within a categorical boundary, are never strengthened and complicated by challenges from other traditions. Having lost the certainties of the old "Voice-of-God" narrative voice, we confront the prospect of a total relativism, in which an account from one universe of interpretation is completely separate from (and consequently is never challenged by) those coming from outside.

If these risks are very generally experienced across the field of history, and have given rise to a wide and complex literature, they have a particular bearing on the writing of the history of Atlantic Canada. A problem of much of the new social history, from the 1960s on, in both its "neo-Marxist" and "post-Marxist" variants, is that it does not allow for the construction of "Canada" itself as a central category of analysis. We seem to arrive at an interpretive impasse. One can either refuse the new insights into the complexity of identity and the manifold determinations of "the social" and issue various calls for "law and order" in the academy; or one can simply dismiss as passé any discontent with fragmentation and learn to live without the existence of any "general readers" for whom one used to write. And if this is the case with "Canadian history", it is perforce (and perhaps even more radically) the case with "Atlantic Canadian history". Why even have a field of Atlantic Canadian history, if "Atlantic Canada" is an empty space upon which we multiply our incompatible and incommensurate stories?

My (very tentative) answer to this impasse in the case of Canadian history is to suggest the abandonment of the metaphor of synthesis and to encourage a strategy of reconnaissance. That is, I think it is neither desirable nor possible to produce a Canadian historical "synthesis" in which every Canadian will somehow see him- or herself as part of a great integrated story. That is a mirage, and it is time to let it go. Rather than beginning with the ambition of synthesis, we could begin with the more modest goal of using both the new tools of cultural theory and the old tools of political narrative and economic analysis to illuminate specific and well-defined issues. In this more problem-centred, bridge-building approach, to "rethink Canada" does not mean to synthesize and integrate all Canadian experience into a universally acceptable account, but rather to map Canada’s logical and historical conditions of possibility as a specific project in a specific time and place.

That project, very briefly, was the project of liberal order, a coherent attempt to install a liberal public philosophy and practice of "peace, order and good government" in northern North America — a programme derived largely from Britain but influenced also by the American example and by North American social realities. Canada was thus a process, not a self-evident thing, and the key to this process was liberalization: the drive to make certain specific and historically contingent assumptions about human beings and their societies natural, obvious, "commonsensical" — in a (loaded) word, "hegemonic". In the absence of any deep geographical or ethnic "foundations", the Canadian state did acquire original and distinctive features, and it gradually came to invent traditions and discourses that were similar to but also distinct from those in other places. Liberal hegemony in Canada was secured against entrenched enemies, difficult challenges, an enormous geography, the pressures of other national projects: each of these in turn would subtly influence the project of Canada. Let us also keep in mind that the liberal order in Canada was ineradicably shaped and changed by those who opposed it; the extension of its hegemony across North America was a dramatic, revolutionary, contested, difficult and contingent process, and it remains so today.

What has this got to do with Atlantic Canada? Everything. A well-traveled polemical path, pursued by both sides in our country’s rather mediocre history wars, has been to lump in "regionalists" with all those other — feminist, working-class, multicultural — pesky scholars who have so fecklessly undermined the old ways of understanding Canada. The issue is surely far more complicated. Should "Canada" disappear as a category of analysis and as the focus of historical inquiry, many important traditions of "regional inquiry" will evaporate right along with it. Contrary to the myth-symbol complex of the old Toronto School, which still casts its shadow across our historiographical landscape, regional scholarship (and "regionalism" as a sentiment) was not summoned forth by the 1960s vogue for "limited identities"; one can surely say that it had arisen along with "national history" from the 1920s on. Much of the drive to write the history of the region, from the days of Harold Innis, S.A. Saunders, D.C. Harvey and J.B. Brebner, was implicitly to fit the region within larger national or continental frameworks. Thus to be lining up the "regionalists" as one more "balkanizing faction" in the history wars seems theoretically and empirically mistaken.

At the same time, from the 1960s on, there were convergences between regional history and social history. Besides obvious overlaps in approach, such as that of working-class historians piecing together "regional patterns", there was a sense in which a regionalist could only sympathize with a skepticism about the nationalist master-narrative, which had so often peripheralized or marginalized topics located in "Atlantic Canada". Moreover, it seemed much more plausible, in certain areas, to theorize beyond one’s community or group-based evidence to a regional level of analysis. In the case of certain ethnic groups, such as Blacks and Acadians, for example, there was an empirically strong case to be made for a regional framework of analysis as one that reflected the defining peculiarities of a particular group beyond the purely local level.

But this means that, in the contemporary civil war within Canadian history, regional historians are left in a most difficult position. On the one hand they share the sense that the "abolition" of Canada as a topic also means the "abolition" of many of the traditional concerns of "the region". On the other hand they would also resist any attempt to return to the days when a small number of middle-class men, resident in Toronto, told the story of the inevitability and goodness of the development of Canada. We should add to this the sheer difficulty of articulating "region" as the object of a social-scientific or historical discourse: "region" as a concept, and "Atlantic Canada" as an application of that concept, were weakly articulated, and they were easy subjects for "deconstruction" even before the term had been invented. One cannot help but notice that most of the papers at this Atlantic Canada Workshop are not about a region called "The Maritimes" or "Atlantic Canada", but are rather about the fit between international conceptual frameworks and the experience of "place".

Ernest Forbes’s brilliant dissection of the way in which the "frontier myth" in Canadian history cast the Maritimes in the role of the backward, stagnant "Other" to the vibrant, expanding west was a pathbreaking piece — an anticipatory act of deconstruction — which should eventually call forth many more such attempts to piece together the way in which "region" has functioned in the imagined Canadian community.[2] The person undertaking such a deconstructive reading today would be more challenged, since the evaporation of "region" as a category in the writing of much Canadian social and cultural history means that one must dig, not for candid illustrations of regional stereotyping, but rather for a more subtle language of exclusion and hierarchy.

In Mariana Valverde’s The Age of Light, Soap and Water: Moral Reform in English Canada, 1885-1925,[3] for example, we are introduced to the "Toronto-Ottawa axis", and to an "English Canada" that can be meaningfully extended, without significant institutional breaks, back from today to the 1920s. In this pioneering work, which deservedly won widespread praise for its courageous introduction of a new methodology, one experiences a certain vertigo of placelessness. We find ourselves somewhere in Ontario, within an "axis"; it seems that when we grasp a part, we grasp the (vaguely limned) whole. This ambitious use of the trope of synecdoche is highly characteristic of the historiographical temper of our age, which is simultaneously drawn to the very particular — this speech, this metaphor, increasingly this murder trial — as an instance of the massively general — patriarchy, chivalry, racism and so on. Yet how do we know that we have grasped such abstract forces without a sense of their typicality within a country or, more feasibly given the constraints under which we all work, within a region? It could well be that the writing of Atlantic Canadian history has been skewed by being so placed within this axiological framework. What if, for many purposes, the more pertinent "axis" for Atlantic Canadians was the Boston-Halifax axis — for we have only begun to explore the fact that, from the early 19th century, the region lay in the shadow of the second most densely industrialized society on the globe? Where were we? Perhaps not long, and not decisively, in "Canada" as the Ontario social historians have so insistently, but narrowly, defined it.[4]

Canadian religious historiography, to suggest a further example, has deeply internalized a model of the "evangelical century" and of the elaboration of the "social gospel" at the edge of the industrial frontier: when a clergyman speaks in Toronto, his voice echoes throughout the Dominion. Canadian intellectual historiography has been seized with the rather pleasing illusion — but it is an illusion — that words spoken by professors at Queen’s University resound from coast to coast. It no doubt reflects my own bias when I say that Canadian working-class history, so much of which has been shaped at the University of New Brunswick, Dalhousie and Memorial, has been far more aware of region and far more "inclusive" than other fields. Nonetheless, even here, one is often presented with narratives of the rise, consolidation and restructuring of "the Canadian working class" which work to naturalize rather homogeneous time-lines according to an Ontario logic (and when it comes to the startlingly different Nova Scotia evidence, treats this as simply "exceptional" or "surplus"). Reading beyond the discipline of history, one finds many additional examples of the ambitious use of the "trope of synecdoche". Linda Hutcheon in As Canadian as . . . possible . . . under the circumstances developed an interesting case of a "doubled Canadian voice", that of the forked tongue of irony, by explicitly acknowledging that she was referring to the "pluricultural entity known (somewhat reductively) as English Canada". Then, ironically enough, in a vintage case of nationalist reconscription, she retrospectively rebaptises Thomas Haliburton as a Canadian in her exploration of a single "English Canadian culture".[5] Among political scientists, particularly within the vast industry which today probes the never-ending "constitutional crisis", a well-established trope has been that of the "dream undone", the Canadian "nation" founded in 1867 undone by the pettiness of the provinces and racked by regionalism, not to mention by the Judicial Committee of the Privy Council, whom Eugene Forsey referred to as the wicked stepfathers of Confederation who took up the task of "turning the constitution largely inside out". (That more recent scholarship has shown that Ontario politicians themselves undid the "dream", and did so precisely in the name of a more powerful and equitable realization of the liberal project, is surely somewhat inconvenient). Or we can turn to Jeffrey Simpson’s well-regarded volume on patronage, with its breathtakingly simplistic view of a west-to-east pattern of ascending political sleaze in Canada, an almost uncanny repetition of precisely the regional stereotypes Ernest Forbes so deftly punctured.[6]

My point, then, is that we do not confront a situation today in which "regionalists" are on one side or the other of Canadian history’s civil war. We are caught in the middle. And in many respects, one regrets to say, many social and cultural historians have gone Donald Creighton one better: at least the commercial empire of the St. Lawrence was explicitly and candidly about a process of extending imperial control, whose ethnic and geo-political assumptions were out in the open. With this new literature, which wears its socialist populism on its sleeve, one often feels one is reading a new and improved version of the same old script, this time with a cast of millions.

I could, like many other regional historians, go on in this vein for hours. I am sure many of us in this room have well-stocked backpacks full of examples. No doubt many of us have voiced like complaints in public or private. What we may wish to reflect on is how little impact such objections, even when well-founded, seem to have. They are all so easily brushed aside, as so much "detail" that disrupts the crucial task of synthesis; or as sheer negativity; or as parochialism. The deep historiographical naturalization of "Canada", its boundaries, power relations and institutions, stems in part from the traditional proximity of scholars to the state, which (like any state) rewards its friends and not its enemies. It also flows from the enormous wealth of Ontario’s cultural apparatus, which is still able to fund an increasingly disproportionate number of research scholars in graduate schools, and from the cumulative impact of the pragmatic decisions these scholars must often make when doing research at Ontario universities (better to drop costly plans for long-distance research and concentrate on sources closer to home). As a consequence, the monographs spill out in an immense wave from Ontario and Quebec, quite out of proportion to Central Canada’s demographic or cultural weight over the long run of Canadian and British North American history.

But beyond such practical considerations lies the strength with which Canadian scholars have embraced "Canada" as their unproblematic and hence, in a way, unexaminable "imagined community", which they feel called upon to celebrate, even if only in "reading back" and eternalizing beyond the 1940s such concepts as nationalism, citizenship, liberal democracy and so on. The country’s actual status as a British Dominion recedes in memory; the territorial extension of a liberal empire to the High Arctic, to Newfoundland and Labrador, to the distant reaches of the North Pacific is presented teleologically as the obvious and inevitable unfolding of a destiny. Such question-begging assumptions not only demean and marginalize those — such as the ultramontanes and nationalists in Quebec and the resistant First Nations — who did not have the good grace to realize the inevitability and goodness of the Canadian project; but it also trivializes the project itself, rendering both the obstacles it confronted and its often ingenious hegemonic strategies invisible. It is one thing to demand of Canadian historians that they ought to include a few more Atlantic Canadian examples in their books — that, say, Ramsay Cook and R. Craig Brown, in their classic Canada: A Nation Transformed, might have found room for just one chapter on the Atlantic Region as a logical complement to their chapters on other regions.[7] It is quite another thing to demand of Canadian historians that they change the ways in which, for six decades, they have "narrated the nation", and to urge that they have, in many respects, mistaken imperial realities for national dreams. In archaizing a novel and important new state-form, they have, ironically, robbed it of much of its contemporary interest.

Where do the concepts of "region" and "regionalism" fit into this reconceptualized approach to Canadian history? Quite centrally. But let us first consider what approaches to "region" and to "regionalism" would not fit, at least not as they are conventionally treated, within this notion of Canada as a relatively new nation and as the continuation of a 19th-century liberal project.

First, one would largely dispense with attempts to define "region" in natural essentialist terms. That it has often been so defined is a matter of record. The notion of the "Maritime Provinces" — when was this interesting term lodged in Canadian discourse? — seeks to define the region according to its intrinsic essence: the sea. This has been a favourite staple of texts and tourist-trade books; and on occasion the insistence that all "Maritimers" should have a mystic bond with the sea has had its humorous side. (Alan MacEachern, in his thesis on the region’s national parks, in itself a wonderful example of the invention of Canada as a "natural entity" in the 20th century, documents various instances in which politicians in the region designated park-worthy settings in the interior of provinces, only to encounter the objection of the Ottawa planners: by definition, and as a matter of course, a national park in the Maritimes must be by the sea! And, until the 1960s, all of them dutifully were).[8]

Without recourse to ultra-relativism, one can say that the boundaries of the region have shown a disquieting tendency not to behave like permanent fixtures on the landscape. They are socio-political boundaries. Even the definition of the region provided in the first issue of Acadiensis, surely the most authoritative place to look for certainty on this issue, was somewhat amorphous:

Devoted to focusing regional awareness, [this] journal will concentrate upon Atlantic Canada, but will include within its geographic scope not only the Maritime Provinces and Newfoundland but also Gaspesia and Maine with further extensions into Central Canada and Northern New England when these seem relevant. Regional in its approach, Acadiensis welcomes American studies that enrich our understanding of communities on both sides of the American-Canadian border. European studies that impinge upon the development of the Atlantic region are within the scope of the journal.[9]

There is, one notes, no actual definition of the "region" designated by the phrase "Atlantic region", and there is the implicit suggestion that the Gaspé and Maine are at least within its penumbra.

In the actual practice of the journal, however, "Atlantic Canada" has been treated rather differently than this implied definition, both more narrowly and more broadly. It has been treated more narrowly in that the largest single part of the Atlantic Provinces has gone virtually undiscussed: Labrador, whose land-mass is greater than that of the island of Newfoundland and, indeed, that of the other provinces combined, has almost never been discussed in the pages of the journal, and in fact figures very little, if at all, in the imagined region most "Atlantic Canadians" think they live in. Defined in terms of the territorial claims of an expansionist Canadian project, the region’s borders have shifted quite considerably, aided in part by the Judicial Committee of the Privy Council (whose settlement of the Quebec/Labrador boundary issue may or may not hold up should Quebec attain independence), and in part of course by the extension of Confederation to the tenth province in 1949. It was only then, I suspect, that the phrase "Atlantic Canada" came into common usage, displacing such older terms as "the Lower Provinces" (very common in the 19th century). What very likely began as a census category has since been naturalized as a regional inevitability and a focus of identity. It is for this very reason — that the term was an Ottawa invention, which imposes a top-down homogenization on historically distinct experiences — that the name has also been resisted. (On the other hand, in Ontario, one often encounters a nomenclature that lumps all four provinces together as part of the "Maritimes"). There is also the question of Canada’s sovereignty over coastal waters, which did not extend the boundaries of the provinces, but certainly changed the futures of many who lived within them. Attempts to define the region "commonsensically" in terms of widely-acknowledged and shared geographical features, stable boundaries and "ethnic essences" do not appear very promising in the light of a history of such boundary revision and in the face of a population ranging from the Inuit in the far North to the New England-descended population of the far South. Tracing this "naturalization" of the region in a more sustained and historical way could provide a researcher with some fascinating insights into how humans imagine geographies and, over time, turn these imaginings into real lines drawn upon the land. It would also provide insights into the ways in which this "natural definition of region" was the complement of, perhaps even the response to, simultaneous Canadian-nationalist attempts to find a natural warrant for the nation-state in the Canadian Shield (canonized by the Group of Seven), the St. Lawrence River Valley or (more abstractly) in the east-west axis that Canadian nationalists hold so dear.

A second, somewhat different if related take on the "region", is one we might term the "structural-functionalist" one. By this we designate not one theory or approach in particular, but a collection of approaches that bear a family resemblance to each other. In such discussions, the "region" is defined as the subordinate and inferior term of a binary opposition: the "periphery" to a "centre", a "hinterland" to a "heartland", an "internal colony" to a "metropole". The Atlantic Region comes to be defined largely in terms of its structuring absences: its lack of a metropolis, its lack of domestic pools of capital, its lack of a well-developed industrial base, its lack of a "developed class structure".

Heartland/hinterland models, developed particularly by such scholars as Graeme Wynn and Larry McCann, have offered tools to describe the evolving regional geography of Canada. In this non-Marxist version of the theme, regions were seen to interact with each other, economically, socially, culturally and politically. Both heartland — generally meaning the Windsor-Quebec City corridor — and hinterland are distinguished by the functions they play within a more general structure: the heartland creates a demand for staple commodities, supplying the hinterland with "capital, labour, technology, and entrepreneurship, those factors of production which are so essential for the initial growth and sustained development of the hinterland". And because the source of demand rests at the centre, the periphery is dependent on the centre not only for supplies of capital or technical expertise, for example, but also for patterns of living, social organization and well-being. The theoretical implication is clear: power is mainly the preserve of the heartland.

The Marxist variant of the theory bore some resemblance to this theorization. It too offered a "polarized theory of growth", underlining the unevenness of capitalist development and the extent to which it required areas of high unemployment to provide the "light infantry" of capital. This idea was articulated in Restructuring and Resistance, a book edited by Bryant Fairley, Colin Leys and James Sacouman, and in many respects a continuation of the discussion advanced earlier by Brym and Sacouman’s seminal Underdevelopment and Social Movements in Atlantic Canada:

Capitalist expansion nationally and continentally has been accompanied by crisis and collapse in the Atlantic region’s economy. The offsetting factor has been the growth of the welfare state, a creature of the redistributive fiscal powers of the federal government. The Atlantic region gradually became a dependency, to the point where economic, social and political stability now all depend on the flow of federal dollars. With the exception of a few multinational corporate enclaves, capitalism within the region is itself largely dependent on state subsidies and transfers, and on the aggregate demand generated directly or indirectly by state expenditures on goods and services. . . .[10]

These structural-functional realities entail certain consequences for classes within the region:

Petty producers and workers in primary industries have occupied a central place in the struggle against economic decline in the Atlantic region largely because of the importance of their industries to what remains of the regional economic base. To some degree, the pre-eminence of primary industries is symptomatic of the underdevelopment of the region: i.e., of the failure of both local and outside capital to generate significant levels of secondary manufacturing activity. On balance, however, these industries are key to the regional economy because they enjoy certain comparative advantages relative to competitors nationally and internationally. The problem of underdevelopment is therefore in large part a function of the failure to build on this base.[11]

It would take me far beyond my remit to go into these paradigms in detail; in the latter case, the "development of underdevelopment", "dependency" and "dependent development" approaches sustained a large and heterogeneous literature, down to about 1990, that really calls out for a major interpretive monograph. My impression — and I stand to be corrected — is that over the past decade this approach has become less prominent. It is certainly noteworthy that the word "underdevelopment" does not figure in any of this year’s conference presentations. I would hazard the speculation that both approaches have come to seem dated by the very air of structuralist scientificity that once gave them credibility. In the case of heartland/hinterland, it did not appear that the paradigm could yield consistent results. It was conceded that, because hinterlands shipped many staple commodities directly to foreign customers, the heartland depended on the periphery; but once this concession was made, the identity of the polarities described by the theory seemed to be unstable.[12] As for the second, it underwent volleys of theoretical and empirical critique from within as well as outside Marxist circles.[13] (For many, its diagnosis of a process of capitalism requiring the permanent exile of a global periphery to conditions of dependence was undermined by the emergence of new economic powers in Asia).

For my part, I still find this literature offers us compelling metaphors and descriptions for real regional problems, without perhaps giving us specifically focused interpretations and explanations. Additionally, I would venture to wonder if there was not a certain tendentious economism and reductionism in certain elements of this tradition. This was particularly evident in Restructuring and Resistance, when it came to discussing the position of subaltern classes outside the "classic proletariat": "In general, the question was raised of how the political potential of any social movement of this kind is to be assessed. The fact that it enjoys widespread support from small independent producers (farmers, fishers, lumber producers and the like) does not necessarily mean that it is actually or potentially anti-capitalist, let alone socialist. It is at least as likely that it will be populist, i.e. defending essentially pre-capitalist values in the name of ‘the people’, but having a very garbled class base, lacking a clear political analysis, and hence being prone to come to nothing (or even to come to serve capitalist purposes)".[14] One pauses over that word "garbled", especially as applied not to a discourse but to a base, and indeed over the entire passage, which conveys vividly the sense of the Marxist theoretician, at the control panel of history, engaged in a constant assessment of which class can be considered the bearer of historical reason, listening for correct answers and hearing only things which are unintelligible or devoid of lasting political potential.

The themes of dependency, underdevelopment and metropolis/hinterland interaction have been descriptively rich, and I have been complicit in their development myself. I think, however, that the next step for this literature is to undergo the rigorous analysis of concepts, the de-ontologization of its categories and the sharpening of its sense that determinist metaphors can force both analysis and politics into an impasse. It is time for such discussions to become more historically specific and contextual. Consider, for example, how many times we have heard of the catastrophe of the 1920s and 1930s. True enough, in its way: the statistics are there, and no one would argue that the coal and steel industries were not crisis-ridden. Yet think also of the great transformation of the regional economy in precisely this decade: the rise of Irving, the expansion of pulp and paper, the triumph of tourism, the aggressive entry of the state into everything from road paving to hotel management. This would not seem to be a straightforward story of a hinterland deprived of agency, but a more complicated one, one of whose important themes is the rise of a new service capitalism. By focusing only on the dire master narrative of the region’s fall into dependence, we have missed many interesting other stories about enterpreneurs and others who saw potential in crises.

A third resource upon which one might draw for a reconsidered late-20th-century defence of both "region" and "regionalism" is, of course, the tradition of Canadian national historiography itself and its attempt to elaborate a myth-symbol complex adequate to the needs of a people undertaking the (possibly impossible) task of inventing themselves as a nation in the mid-20th-century. Insofar as this has involved the effacement of what is distinctive and rich about the history of the Maritimes, and later Atlantic Canada, this "nationalizing" historiography is simply to be resisted. When the region and its people are "Othered" in this nationalist historiography, when they are reduced to quaint, patronage-ridden, backward obstacles or simply seen as a "problem of regional disparity", we should query these discursive "terms of trade", this reduction of what we know to be a rich complexity to one inferior term in a binary opposition. These gestures of superiority, on behalf of a metropolis that never has been and never will be accorded the cultural prestige of a true metropolis are simply gestures of its weakness. For it is not at all clear that either Montreal or Toronto has ever really suffused the region with new ideas, with a genuine hegemonic presence involving both consent and domination, philosophical inspiration and political pragmatism, or that in this much more complex cultural sense either metropole, before the 1930s, ever served to integrate the Atlantic Region into a more general socio-political project.

Within the project of inventing a mid-20th-century Canada, Atlantic Canadians have often been the most grounded, cogent and sustained Canadian nationalists. They have invested lives and hopes in this experiment, and without them "left liberalism" as a Canadian option would have been infinitely weaker. They number, uncredited and unstudied, as among the most articulate and important exponents of a general welfare state (one thinks not just of Norman McLeod Rogers, but also, at least in the 1930s, of Angus L. Macdonald; and not just of William Coaker’s detailed manifesto for the fishers, but also of J.R. Smallwood’s admittedly inconsistent embrace of a "developmental liberalism"). Were one arrogant enough to suggest a new agenda for regional studies, one leading item would surely be this: to explore the remarkable extent to which Atlantic Canadians built the Canadian liberal project, and continue to sustain it; the extraordinary depth and subtlety of the Atlantic political tradition, generating organic intellectual after organic intellectual of the progressive liberal order. In telling this fundamentally important story, the notion that these are simply "peripheral" ideas from "peripheral figures" is threadbare. From the 1930s (and perhaps building on older traditions of responsibility) an articulate Atlantic Canadian critique of an unrestrained neo-liberalism was the premise of much that is called "regionalism". Nourished by a grounded sense of the power and the centrality of "community" and "solidarity" as counterweights to liberal individualism and the market, drawing from the immense power of something lived rather than something merely studied, these Atlantic critics of modernity seem to have had an inner strength, a sense of being right. Often when we read accounts of underdevelopment, patronage and dependency we come away with a sense of the sordidness and triviality of Atlantic politics; it is often so, too, when we read certain works of political scientists, who dote on the region’s corruption, venality and conservatism. But these somewhat trivial treatments miss a good deal of what is distinctive about regional politics. From this region, time and again, has come a subtle, complex, communitarian critique of the liberal order. This is in fact one of the region’s defining historical characteristics.

There is also the socio-cultural sense of region. Again, let us begin with refusals. The reduction of Atlantic Canadians to a species of Folk, to "unspoiled peasants", and the reduction of the region to a "playground", a therapeutic space wherein the tired businessmen of Canada could recover their energies: these were all rather unsubtle and reductionist ways of thinking "region". To this day, and especially in Ontario, one repeatedly encounters the primitivist stereotype of the region, now "Othered" less condescendingly, yet in ways no less objectifying and demeaning than the stereotypes which Forbes demolished. Insofar as regional culture has meant the generation for the Tourist Gaze of reassurances, therapeutic spaces and Folk cultures, it needs — on ethical and political grounds — to be resisted.

When we read accounts of the "natural region", with its essences of sea, flora and fauna; when we read structural-functionalist accounts of the "dependent and underdeveloped region", with their sense of finality and objectivity; when we read of the "political region", that space of patronage and small deals in small places; when we (or our impressionable students) read of all this, we perhaps inevitably feel . . . not angry, but depressed. I remember teaching Atlantic Canadian Sociology at Saint Mary’s University in the early 1980s. We worked through all the readily available texts on underdevelopment and dependency. Many of these writings worked to create a sense of finality; every moment of resistance was autopsied, and when the autopsy was concluded, the verdict was clear: the region stood condemned. I felt that, at the end of the class, there was a palpable sense of "no option" in the air: no option, that is, but to leave. Not what I had intended, not what the accumulated weight of critical theory and our "dependency readings" had intended: but there it was. . . .

Looking back, I think I would said to my despondent students that, yes, there is much that speaks of determinism and of finality here. There is also much that speaks of many frameworks — heterogeneous and various, and yet somehow, in ways I would find hard to put into words, also connected to each other — that have been enormously creative. That movements of resistance, more difficult to create in the Atlantic environment, have also, somehow, meant more: that they have echoed, as inspirations and memories, through time in a way one does not find in many other places (and certainly not in "developed" but historically forgetful Ontario). That there is much, much more to this place and this time than an inability to live up to the model of Other Places. That even in highly compromised forms, there is a space of irony, and the possibility of resistance. One thinks of the extraordinary awakening, happening just as so many of us were immersed in "dependency", of independent voices and new cultural forms in the region.

Perhaps this will then be the fifth and final meaning of "region" in Canadian history. It is taking shape before our eyes. A new history of politics — focused on political theory, ideological development and state-formation — is reshaping our understanding of the Canadian experience; and the Atlantic Region will likely play a central role in this research programme. There is much that seems regionally distinctive and unifying about the emergent resistance of the region to the totalitarian logic of global neo-liberalism. I well remember telling my disbelieving sociology students at Saint Mary’s that, in the 1910s, this region was once a by-word for social activism; one spoke, in Central Canadian newspapers, in tones of disparagement about the "Bolsheviks" of the "East". To outsiders, it often seemed an unpredictable place, where new ideas bubbled up, unannounced, yet — when we think about them in retrospect — deeply rooted in their place. So perhaps this will be the task that those in quest of region in the 21st century will set themselves: Why was this region so bounded by economics and so limited by geography nonetheless such a hot-bed of new ideas? Why did it, against all the prognostications of neo-liberal economists and political scientists, become the heartland of the (old but persistent) dream of a new democracy? Perhaps therein lies the final sense of what "region" — this "region" we construct in the present and in our heads, to capture the elusive patterns of the past — may come to mean in our work and in our practice.

IAN McKAY