Roberts’ “Tantramar Revisited”: Another View

by David Jackel

     All serious students of Charles G.D. Roberts’ poetry will be grateful for William Strong’s article, in an earlier number of Canadian Poetry, on “Tantramar Revisited.”1  Strong’s stated intention was “to expand upon the necessarily brief examinations of the poem” offered by earlier critics in order to show that it “attains to a fine and complex unity of form and language, imagery and thought,” and “is, indeed, . . . Roberts’ poetic masterpiece” (p. 26).  The resulting article is a valuable one, in many respects the most detailed and coherent study of “Tantramar Revisited” yet to appear.  One may admit this, however, and still argue that the poem requires further critical discussion.  In particular, I think that the words unity and masterpiece need both explanation and qualification, and that the poem’s place in Roberts’own development and in its Canadian context needs to be better understood.  No more than Strong do I intend to devalue the poem’s high reputation.  My intention in this paper is to examine some of the critical confusions that have grown up around the poem, to point out the difficulties that remain unresolved by Strong’s article, and to suggest some resolutions of these confusions and difficulties that will provide a firmer basis for evaluation of the poem.

     I begin by discussing the formal elements of the poem — its rhythm and structure — in order to demonstrate the extent of Roberts’ control over his materials.  The imagery and themes of the poem raise questions about Roberts’ personal situation that I have tried to answer by placing the work more firmly in the context of its time.  A reader of my essay will benefit by having a text of the poem at hand; the version in W.J. Keith’s Selected Poetry and Prose is both reliable and accessible.


     The verse-form, or metre, of Roberts’ poem has been the subject of much confused discussion, beginning with James Cappon’s assertion that “Tantramar Revisited” was written in “the not very pure form of the modern hexameter,”2 a form made popular by the influence of Longfellow’s Evangeline.  The possible influence of Arnold and Clough on Roberts’ choice of verse-form is also noted by Cappon, whose comments on Roberts’ handling of his metre seem at odds with the generally high praise he accords the poem.  The form is said to be “high but somewhat artificial,” the verse has “an exotic character . . . , which after all is a bar to the highest qualities of expression,” and it is “a strong compelling mould which is apt to draw the poet into iterations and to carry him further than he wishes at one time while reining him up unduly at another.”3  If Cappon can make these criticisms, and go on to add that Roberts uses the hexameter “with some freedom and naturalness” but at the cost “of some rough lines, lines overloaded with awkward spondees or technically impure and sometimes falling out of metre altogether,” then he has undercut his own judgment that the poem is “a true whole” — unless we can somehow argue (implausibly, I think) that metrical lapses of the kind found by Cappon should not affect our assessment of the poem. Cappon’s discussion of the metre makes one additional point: that Roberts uses a “pentameter variation . . . , following Clough’s example in Amours de Voyage.  It is designed of course to afford some relief from the monotonously majestic stride of the hexameter and allow the poet to escape into plainer cadences.”4  Again, this presents a problem; monotony, even when majestic, is not a quality we associate with fine poetry.  Cappon’s examination of the metre is far from thorough, and seems to raise more problems than it solves.5

     Elsie Pomeroy, in 1943, took up the matter more seriously, challenging what she termed Cappon’s “erroneous” description of the verse-form and making her own description the basis for high praise of Roberts’ craftsmanship:

The difficult metre, with its alternating hexameter and pentameter lines, is particularly noteworthy, for rarely has this elegiac measure, so popular with Ovid, been used in English.  Clough wrote his twelve-line introductions to the divisions of “Amours de Voyage” in this metre; and Swinburne dallied with it occasionally and always briefly in “Evening on the Broads”; but no poet, writing in English, has used it with such freedom and sustained precision as Roberts.6

The claim that Roberts has employed the “elegiac measure” is clearly stated, but there is no supporting analysis.

     Desmond Pacey’s considerations of the poem do not clarify matters.  In Ten Canadian Poets we are told that the verse-form “is not original — it is virtually identical with that of Longfellow’s Evangeline — but it is perfectly suited to the theme of nostalgic remembrance, and it is handled with masterly ease.”7  The position adopted here is close to Cappon’s, with the important difference that Pacey has no reservations about Roberts’ metrical skill.  In his last essay on Roberts (1961), Professor Pacey offers a slightly different description of the verse-form:

The poem is written in the classical elegiac metre, in alternate dactylic hexameters and pentameters which had been recently employed by Clough, Arnold and Longfellow.  It was probably Longfellow’s Evangeline that provided Roberts with his metrical model, but the charge of derivativeness has little weight if, as I believe is the case here, the imitation surpasses the model.8

There follow some interesting comments on “Roberts’ power to modulate the movement of the rhythm to suit the material with which he is dealing,” but these comments cannot disguise the fact that Pacey’s general description of the metre amounts to nothing more than a combination of the conflicting views of Pomeroy and Cappon, not a resolution of the conflict.

     Later writers, with the exception of William Strong, have either ignored the verse-form or referred to it in very general terms.  John Ower notes the “marked and regular rhythm,”9 and W.J. Keith comments that “the verse-form is reminiscent of both Longfellow and Swinburne.”10   Robin Matthews seems to find Keith’s view unacceptable (the poem “could never be confused with a production by Longfellow or Swinburne”),11 but he does not say why.  Although Strong argues that Roberts has deliberately chosen “to echo . . . the verse form” of Evangeline, he also claims that the poet “first opens with and then plays against a hexameter norm” (p. 27).  Like other critics he praises Roberts’ metrical skill, but he also notes — correctly, in my view — that Roberts has varied his metre by “playing against it the rhythms [my italics] dictated by the verbal sense and the reading voice” (p. 28).  The point is a good one, but it still does not provide us with a sufficiently accurate description of how Roberts has handled the verse-form of his poem.

     The claim that “Tantramar Revisited” has any significant metrical resemblance to Evangeline can, I think, be disposed of fairly quickly, by putting four-line passages from each poem side by side.

Dikes, that the hands of the farmers had raised with labour incessant,
Shut out the turbulent tides; but at stated seasons the floodgates
Opened, and welcomed the sea to wander at will over the meadows.
West and south there were fields of flax, and orchards and cornfields. . . .

. . . . . . . . . . . . . . . . . . .

Skirting the sunbright uplands stretches a riband of meadow,
Shorn of the labouring grass, bulwarked well from the sea,
Fenced on its seaward border with long clay dikes from the turbid
Surge and flow of the tides vexing the Westmoreland shores.

In the first passage, from Evangeline, we see that Longfellow is indeed employing what Cappon called the “modern hexameter”: the basic unit is the dactylic foot, each line has six feet, the concluding foot is always trochaic, and there are occasional substitutions of trochaic feet for dactylic ones elsewhere in each line.  Whatever we may say of Roberts’ metre, he is certainly not copying Longfellow.  He does employ dactylic feet, but there are more frequent trochaic substitutions; in some lines of the poem, in fact, the dactylic metre almost disappears — in “Miles on miles of green, barred by the hurtling gusts” there is only one dactylic foot.  Of greater importance is Roberts’ arrangement of his lines as couplets through the regular alternation of unaccented and accented syllables in his line endings.  There is nothing like this in Evangeline, where we find an unvarying succession of trochees.

     Pomeroy was correct to call Cappon’s description erroneous.  Yet her own is no more accurate.  The classical “elegiac measure” with its “alternating hexameter and pentameter lines” was a quantitative metre; each vowel sound had an agreed-upon quantity, long or short, dependant upon the time involved in its utterance.  The regular form of the elegiac couplet would, therefore, appear as the following arrangement of long and short syllables:12

scan.gif (1468 bytes)

Because two short syllables were considered equal to one long, a spondaic foot could be substituted for a dactylic one without disturbing the measure.  Yet even if we allow for such substitutions, we cannot argue that Roberts was writing quantitative verse.  Difficulties begin with the first four lines.

scan2.jpg (32184 bytes)

The third foot of the first line is incomplete, and we cannot rectify matters by making “come, and” a spondee because “and” has already been defined by the metre — in the first foot — as a short syllable.  Classical measure does not permit such arbitrary alterations of quantity as would be required to regularize the line.  A similar problem occurs in the fourth foot of the second line.  “Sorrow,” in line three, might be read as a spondee, but only in defiance of normal English accent.  What we might do to regularize the second foot of the fourth line I do not know, and later lines in the poem produce even more serious problems:

scan3.jpg (7956 bytes)

We surely cannot argue that the first two feet of this line are spondaic.

     If we argue that Roberts has adapted quantitative measure to English poetry in the customary fashion, by substituting accented syllables for long ones and unaccented for short, we are no better off.  Although the line just quoted does have six accents, it is lacking three unaccented syllables.  Other lines would give us similar difficulties:

scan4.jpg (8202 bytes)

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

scan5.jpg (8043 bytes)

If we accept Pomeroy’s description then the first of these lines is a pentameter and the second a hexameter; yet the first is lacking three syllables and the second four.  The second line is, as well, predominantly trochaic.  There would seem to be some basis, then, for Cappon’s statement that Roberts has written a number of lines “technically impure and sometimes falling out of metre altogether,” and this in turn would lead us to question Pomeroy’s praise of the poem, founded as that praise is on Roberts’ mastery of the elegiac measure.

     I think that we would be better off to turn our attention from Roberts’ supposed metrical models to his practice in the poem itself.  Roberts’control of the rhythm of his lines deserves high praise, and it is on rhythm rather than metre that our attention should be focused.  The poet has adapted, not attempted to copy, the classical elegiac measure: he has employed six stressed syllables in each line, and retained the couplet effect by alternating unstressed and stressed syllables in his line endings.  The second line of each couplet is, in addition, usually — but not always — slightly shorter than the first.  If we search for a metrical norm we find the dactylic foot most often used, yet Roberts makes so many trochaic substitutions — and even a few iambic ones — that it seems more appropriate to see the lines as consistently having six stresses, with a varying number of unaccented syllables used (in combination with punctuation and the deliberate arrangement of vowel and consonant groupings) to accelerate or retard the movement of each line.

scan6.jpg (33566 bytes)

Here, in the passage where Roberts gives us the most vivid, and the culminating, recollection of the energy of life on the Tantramar, the movement of the lines is slowed by internal punctuation, three strong end-line pauses, and some consonant groupings that require careful enunciation — the transition from “windlass” to “groaned,” for example, which is further retarded by the intervening comma, and the similar effect obtained in the next line with “ponderous lengths, uprose.”  In other passages Roberts speeds up his lines by avoiding internal punctuation and using alliteration: “Dotting the broad bright slopes outspread to southward and eastward.”  The number of accents remains the same, however; each line has the same ordering principle behind it, yet there is sufflcient scope for variation to prevent the order from becoming tyrannical.  The couplets also provide an ordering principle, with variety provided by the occasional use of run-on lines and by different sentence lengths (of two, four, and six lines; there is also one half-line sentence).  One other aspect of Roberts’ handling of rhythm should be noted: the way in which he ends his poem by adopting in the last half of the last line a firmly iambic metre:

scan7.jpg (7118 bytes)

The metre is not what we have come to expect in the course of the poem, and its use at this point brings “Tantramar Revisited” deliberately, and with seeming authority, to a conclusion.

     There is reason, then, for praising Roberts’ poem on the basis of his handling of its rhythmical elements. We must admit that on a few occasions he is drawn into what Cappon termed “iterations” (as in “Dips from the hill-tops down, straight to the base of the hills,” which seems both iterative and tautological), but in most lines Roberts is admirably in control of his material, and the evidence of that control would support a more detailed analysis than I have offered here.  It would be a service to the poet’s reputation, therefore, if we were to stop associating the verse-form of “Tantramar Revisited” with anything written by Swinburne, unless our intention is to emphasize, in the case at least of this poem, Roberts’ superiority.13


     Although there has been some confusion in the discussion of Roberts’ metre, the subject has at least been much discussed.  The structure of the poem has been discussed in some detail only by Desmond Pacey and William Strong.  Professor Pacey offers the following description:

. . . . the poem has a definite and satisfying structure: we begin with ten lines in which the poet describes his own emotional situation; then follows the long middle section in which the poet so vividly describes the Tantramar region; and the poem concludes with ten lines which bring us back to the poet.   The middle section itself has structure: we are led out from the nearby houses of the fishermen-farmers of the Fundy coast, across the grassy marshes and the sea to the far distant hills, then back again across the water to the nets on the shore and so to the men who use them.  The whole process, whereby the eye is led by a series of highlights and curving lines in a spiraling pattern, is very similar to that of a painting on canvas.14

Elsewhere Professor Pacey makes substantially the same point, adding that Roberts “completes the design by bringing our attention back to himself, and his awareness that even here, if he looked closely, he would find the hands of chance and change at work.”15

     One cannot quarrel with Professor Pacey’s statement that the poem has a “definite structure”; and he seems right to insist that Roberts’ command of his materials is evident in the way he controls our “seeing” of the landscape in the poem.  His assertion that the poem has a three-part structure is, however difficult to accept when we observe that Roberts himself has offered us a five-part structure, with the sixty-four lines of the poem grouped in the following fashion: 24-12-14-4-10.  Within each part, moreover, Roberts has paid special attention to such matters as sentence length and the pace of his lines, rendering the character of each section appropriate to the material it contains and the sequence of the sections appropriate to the development of the ideas in the poem.  In other words, the structure of the poem is an aspect of its rhythm.

     Strong has noted Roberts’ use of a five-part structure, and has shown how the images and themes are developed through this structure.  It is possible to extend his point by showing how Roberts effectively uses various rhythmic devices to support this development.  The twenty-four line first section of the poem shows Roberts at his most deliberate.  In the opening two sentences, of four lines each, the speaker asserts the pervasiveness of chance and change everywhere but in the landscape before him, and in the following three long sentences — of six, four, and six lines — he surveys that landscape in an orderly manner, beginning with his own “vantage-ground” and working outward to the horizon.  The lines move with appropriate slowness: end-line punctuation keeps the reader from hurrying through the material (there is only one run-on line), and there are strongly-marked mid-line caesuras in thirteen of the twenty-four lines which further retard the pace.  In other cases, although a caesura is not indicated by the punctuation, consonant combinations produce similar effects: the “s” sounds, for example, in “Skirting the sunbright uplands stretches a riband of meadow.”   The second section, of twelve lines, is not similarly controlled.  Here, where the speaker turns from observation of the landscape to recollection of his experience in that scene, the sentences become shorter (four of two lines, and one of four lines) and there is a high proportion of run-on lines (six of twelve).  The mid-line pause, furthermore, is noticeable in only four of the twelve lines.  We see that the poet is emphasizing the speaker’s inability, noted by Strong, to maintain the assurance asserted in the opening section.  Instead of a sense of stability reinforced by even cadences and balanced statements, we are given in this second section an increasingly emotional departure from the present, underscored by broken rhythms and looser syntactical structures.  The effect is particularly evident in the concluding four-line sentence:

Now at this season the nets are unwound; they hang from the rafters
Over the fresh-stowed hay in upland barns, and the wind
Blows all day through the chinks, with the streaks of sunlight and sways them
Softly at will; or they lie heaped in the gloom of a loft.

Earlier in the poem the syntactical unit has generally coincided with the line or the half-line; here we have three successive run-on lines and the lines are broken internally by punctuation which does not, with one exception, occur near the centre of the line.

     The third section, although it begins with the net-reels that the speaker “sees,” is largely a description not of the observed landscape but of the landscape as it will be when afternoon turns to evening, when evening is followed by night, and night gives way to “the live keen freshness of morning.”  Again, it is the speaker’s prior experience in the landscape that enables him to offer this description (the fourteen-line section might be compared, as Strong has suggested, to some of Roberts’ sonnets),16 and the principle of order afforded by the familiar diurnal cycle helps restore, in part, something of the stability evident in the first section.  The proportion of run-on lines (five of fourteen) is less than in the second section, and there are strong mid-line pauses in half the lines.  The sense of order and permanence thus reasserted does not last, however; the third section ends as the second does, with run-on lines and increasing emotion as the recollection of morning leads the speaker back to the net-reels:

Then, as the blue day mounts, and the low-shot shafts of the sunlight
Glance from the tide to the shore, gossamers jewelled with dew
Sparkle and wave, where late sea-spoiling fathoms of drift-net
Myriad-meshed, uploomed sombrely over the land.

The parallel structuring of the conclusions of sections two and three stresses the importance, for the speaker, of the nets and the reels, and the two come together in the fourth section, which is the crucial one.

     The fourth section stands out from the rest of the poem for several reasons.  It is significantly shorter; it contains the poem’s shortest sentence (“Well I remember it all”); it has both mid-line and end-line punctuation; its movement is further slowed by additional punctuation and by awkward consonant clusters.  In addition, the importance of this section is stressed by the repetition of the short opening sentence as the concluding phrase of the fourth line.  Roberts’ speaker is shown here at his most deliberate and emphatic — and yet the poem does not end here, despite the return, with greater force, of characteristics prominent in the opening section.

     The ten-line fifth section which does conclude the poem contains no run-on lines, and so in this respect is more controlled than sections two and three.  It resembles section three, however, in its infrequent use of mid-line pauses (only three in ten lines), the absence of which prevent Roberts from balancing half-lines against each other as he does in the first and fourth sections.  This slight relaxation of control is appropriate to the speaker’s hesitant conclusion, his unwillingness to test his earlier assertion that there is “no change” in the landscape before him.  The arrangement of the three sentences in this section is also appropriate.  Two four-line sentences in which the speaker attempts to justify his inactivity surround a two-line exclamation in which he recalls once again the “old-time stir” celebrated in section four.  If this concluding section does not resolve the speaker’s difficulties, it does at least show that he has achieved some awareness that his earlier confidence was not soundly based. The structure of the poem, and the character of each of its parts contribute as much as his direct statements to our understanding of the speaker’s mind.  Roberts has given as much attention to the larger shape and movement of “Tantramar Revisited” as he has to the rhythms of its individual lines.


     The imagery of “Tantramar Revisited” has not, like the structure, been neglected; nor has it, like the verse-form, been productive of critical confusion.  Cappon, J.D. Logan, Pomeroy and A.J.M. Smith have all praised Roberts (although in rather general terms) for his ability to present a landscape effectively in language.  Desmond Pacey made the same point, at greater length and with judiciously-chosen examples.  For these critics, what we see in “Tantramar Revisited” is noteworthy for its accuracy, its concreteness, and its vividness.  Later writers have not disagreed with this view, but they have focused their attention as well on other aspects of Roberts’ imagery.  W.J. Keith notes the emphasis placed on the “daily, seasonal and yearly cycles,” and he notes, too, the way in which the imagery is used to good purpose: “Phrases and images within the poem themselves recur and repeat, like the pattern of grass and dykes on the marshes.”17  D.G. Jones draws attention to the way in which images drawn from the landscape mirror the movement of the speaker’s mind between “childhood memories of a sustaining activity and the hushed inactivity of the present”; he also suggests that we should note the contrast between “the ordered life of the villagers” and the potentially destructive power of the sea.  For Jones, this contrast (observed but not commented on by the speaker) suggests Roberts’ “increasing inability to affirm the immediate moment” by taking account of “the mutability and irrationality of life.”18  John Ower’s comments on the imagery are in part an extension of the point made by Keith and in part a denial of Jones’ claim that Roberts is withdrawing from life: ;

. . . . the poet’s recurrent allusions to regular rhythmic motions such as the “surge and flow” of the tides, the “waves” made by the wind in the hay, the swinging of the fishermen’s nets, and the “slow-wheeling” of a hawk, suggest that at least in nature there is stasis and order amid apparent flux.  That these cyclical rhythms can in their turn impart a measure of order and stability to human life is apparent in certain of the details which are enumerated by the poet.  There is, for example, Roberts’ memory of the winding of fishnets at evening in which the circling of windlass and reel replicates the seasonal and diurnal cycles by which life in a pastoral community is governed.19

Ower seems right to stress the “cyclical” elements in the imagery which culminate in the fourth section of the poem; Roberts’ handling of rhythm and structure (earlier discussed in this essay) emphasize the importance of this section.  One might note as well that the winding of the nets makes a connection between land and sea which brings together the contrasting images seen by Jones.

     In concentrating their attention on Roberts’ “painting” of the landscape and, more recently, on the landscape as reflective of the speaker’s own attitudes, these critics have neglected to notice that there are contradictions between what the speaker says about the landscape and the imagery in which that landscape is presented.  Strong’s discussion of the poem’s imagery brings out this point very well.  He shows that what we are given in “Tantramar Revisited” is not a “natural” scene but rather one resulting from the interaction of man and nature through many years.  We might also note that the speaker in the poem presents this process ambivalently: the houses are “stained” with time, and yet it is human activity through time which has set them “warm” in their surroundings.  This inability on the speaker’s part to recognize that a good deal of what gives the landscape its charm derives from that same process he deplores raises two possibilities not considered by Strong: that Roberts intends some irony at his speaker’s expense, and that the speaker’s refusal to test his assertion should be seen by the reader as issuing from a failure of perception of which the poet is conscious.

     That there is ironic distance between the poet and the speaker in the poem becomes more than a possibility when we consider an additional set of contradictions, which in this case undercut the speaker’s claim (accepted by D.G. Jones) that there is a clear contrast “between childhood memories of a sustaining activity and the hushed inactivity of the present.”  It is true that the only overt activity we see in the poem comes with the memory of “men at the windlass,” and that the speaker contrasts the “old-time stir” with the “present peace of the landscape.” Yet the imagery associated by the speaker with the landscape shows that he is unable either to present a static portrait or to avoid seeing the landscape, past and present, as intimately bound up with human activity and human feeling.  In the first section we have (I use italics for emphasis) the “labouring grass” and the “hurtling gusts,” and the tides “vexing the Westmoreland shores.”  In the second section the flats are seamed and baked in the sun.”  In the third we have the “gossiping grass,” the “lonesome / Golden afternoon,” the crane which journeys “homeward,” the “winnowing soft grey wings” of the marsh owls, “the live keen freshness of morning,” the “teeth of the dawn,” the “awakening wind,” and the shafts of sunlight which “glance from the tide to the shore.“  It might be argued that Roberts is simply making rather casual use of the pathetic fallacy, but this argument can be countered.  Given Roberts’ deliberateness elsewhere in the poem, a relaxation of control over the imagery would be unlikely.  And, in fact, the increasing insistence on human feeling in the imagery leads us, like the rhythm and the structure, to see the crucial importance of the fourth section.   Here we not only have men at work, we have men who, when their work is done, can return home: the human feelings which permeate the landscape reflect the relationship between the marsh country and the people who inhabit it.

     Understanding this, we can appreciate what is involved in the speaker’s refusal, in the fifth section, to “go down to the marshland.”  The refusal is made because the speaker knows he is no longer “at home” in the landscape; he is an alien (this is reinforced by the word “spy”) who can participate in the life of the region only through memory.  His illusion that life on the marshes is unaffected by change is not merely (as Jones suggests) the product of distance, nor (as Strong argues) a demonstration of man’s psychic “need for illusion as well as reality” (p. 34).  It is truly a “darling” illusion, the offspring, as the speaker recognises, of his own wishful imagination.  The Tantramar community perpetuates itself through time, but the speaker chooses to separate himself from that community and to perpetuate its existence for himself by generating an illusion which can live only in his mind — protected (like a helpless child) from contact with the community.


     In my discussion of “Tantramar Revisited” I have tried to suggest that we must necessarily distinguish between the poet and the speaker.  Roberts is in control of the poem’s various elements — its rhythms, its structure, its imagery — and this control seems intended to show us the speaker’s difficulties in making a proper adjustment between his response to the scene and the essential quality of what he is observing.  If this suggestion is acceptable, then we must argue that the poem is something more than accurate observation of the Canadian landscape colored by nostalgia or elegiac remembrance.  To make such a claim is not to debase the poem’s value but to raise it.  Roberts has invited us to think about his speaker’s attitudes, not simply to indulge in a sentimental sharing of them, and poetry that tries to engage both our intelligence and our emotions is poetry that deserves our attention and respect.  Yet we can praise the poem and, at the same time, note that Roberts has failed to deal completely with the intellectual issues he has raised.  As W.J. Keith has said, the poem ends “somewhat defensively,”20 and D.G. Jones terms the speaker’s final position “precarious” — as if he were poised for action or affirmation but unable to move without losing his balance.21  We might extend these views, and say that the poem simply does not end satisfactorily: Roberts has led us up to a significant question and then turned away from it; the poem is incomplete.  We cannot agree with Robin Mathews, therefore, that “Tantramar Revisited” is “flawless.”22  We can, however, attempt to understand why the poem is flawed.  The reasons are to be found, in part, in Robertst own uncertainty about the relationship between thought and emotion in poetry; his difficulties in this area would have worked against any attempt to resolve the issues raised in the course of “Tantramar Revisited.”  The reasons are also to be found elsewhere, in what we might term Roberts’ “regional” and “Canadian” circumstances; he could not resolve his difficulties because he did not quite understand what they were, but his poem — seen in its proper context — is an important contribution to a definition of those questions about the relationship between the Canadian writer and his society which still properly concern us three generations later.

     In his important essay, “The Poetry of Nature” (1897), Roberts provides us with a clear statement of what he understands poetry to be:

To say that poetry is the metrical expression in words of thought fused in emotion, is of course incomplete; but it has the advantage of defining.  No one can think that anything other than poetry is intended by such a definition; and nothing is excluded that can show a clear claim to admittance [Roberts’ italics].23

In theory, at least, Roberts grants thought a place in poetry; and he makes the significant additional point that thought and emotion must have some connection, and not simply coexist in a poem as separately discussable elements.  Yet later, in the same essay, Roberts deals with poetry exclusively in terms of its affective power.  His remarks show his concern for assessing the quality — or intensity — of the emotion evoked by a poem, but the quality of the thought involved is not discussed.

Nature becomes significant to man when she is passed through the alembic of his heart.  Irrelevant and confusing details having been purged away, what remains is single and vital.  It acts either by interpreting, recalling, suggesting or symbolizing some phase of human feeling.  Out of the fusing heat born of this contact comes the perfect line, luminous, unforgettable, with something of mystery in its beauty that eludes analysis.  Whatever it be that is brought to the alembic. . . ., it comes out charged with a new force, imperishable and active wherever it finds sympathies to vibrate under its currents [my italics].24

As the italicized passages show,  feeling is of central importance; thought is being left to take care of itself.  Given this reluctance on Roberts’ part to take up the question of how poetry affects our intelligence, and his failure to explain how thought becomes fused in emotion, we can conclude that his theory of poetry would not encourage a rigorous attention to the intellectual issues raised by a poem such as “Tantramar Revisited.”   Although we can apprehend the feelings of the speaker, we do not fully understand what motivates those feelings, and Roberts takes us only to the point of wanting to understand and then leaves us there.

     But it is not, I think, Roberts’theoretical difficulties which contribute most to the irresolution at the end of “Tantramar Revisited.” D.G. Jones has suggested — correctly, in my view — that it is to Roberts’ personal life we should look for an explanation, and he refers to the poet’s “personal disillusionment” and to the “gradual withdrawal” evident in In Divers Tones (1886), the volume in which “Tantramar Revisited” was first collected.  Although John Ower does try to offer a counter-argument, his claims are questionable:

The work of art . . . seems to emerge for Roberts as a transcendent synthesis of past and present, movement and stasis, mind and landscape, in which the demons of alienation and change can be mollified at least while creative activity is in progress.  Like Yeats’ “Byzantium” poems, “Tantramar Revisited” may show in its unfolding the creation of an “artifice of eternity,” a “darling illusion” within the compass of which the otherwise insoluble dilemmas of mortality may be overcome.25

Ower’s use of seems and may give his argument a tentative quality, and many readers will probably insist that there is an important distinction to be made between an “artifice of eternity” and a “darling illusion.”  Furthermore, if the demons of alienation and change can only be mollified while the poet is in the act of writing about them then the finished poem is of little value to the reader.  William Strong does argue that the “darling illusion” can sustain the reader as well as the poet (pp. 34-35), yet his claim does not account for the “defensiveness” that Keith finds in the conclusion.  Professor Jones’ suggestion that we look to Roberts’ personal life seems to remain our most useful starting point.26

     I would argue, however, that Roberts’ own circumstances, while important, have not been thoroughly considered.   “Disillusionment” is not the word appropriate to the sentiments expressed in his Alumni Oration, “The Beginnings of a Canadian Literature” (1883), delivered in the same year which saw the first publication of “Tantramar Revisited”:

Literature is not only the revelation of present mood and character, but it also has in its hands the moulding of future character.   The exponent of the present, it is also the architect of the future.  It is an argument never ended, that concerning great men and their times.  One says this man moulded his time; another, that he was the product of his time and his circumstances.  In truth the man and his time act and react upon each other; but the time and the circumstances get rather the best of it, probably.  These unmistakably speak through the man.  But he makes the times and circumstances which shall mould his successors.27

If Roberts is not here naively optimistic about the possibility of human action undetermined by circumstances, neither is he pessimistic about the possibility of influential human achievement through literature.  And the passage is interesting not merely because it invites us to question Jones’ assessment of Roberts’ state of mind; it is also significant because it shows Roberts capable of speaking firmly on the very questions — of time and change — hesitantly treated at the conclusion of “Tantramar Revisited.”  This does not mean that Roberts was confused in his attitude to these questions; the poem, like most good poems, deals more seriously with human experience than do most alumni orations.  In addition, the poem bears directly on the poet’s own experience in a way that the oration does not.  In 1883 Roberts had resigned his position as a teacher in Fredericton in order to move to Toronto and assume the editorship of Goldwin Smith’s new publication, The Week.  It was in that periodical, on December 20, 1883, that “Tantramar Revisited” was first published, and I think it can be argued that the hesitancy at the conclusion of the poem reflects Roberts’ personal sense of dislocation, his uncertainty over whether he could, “far off,” continue to participate in the life of his region.

     The move to Toronto, seen as a context for the poem, perhaps helps to explain Roberts’ repeated reference to “chance and change.”  Change is a prominent theme in the poem, and the critics have given it due attention.  Chance is mentioned, but not illustrated by specific examples, and so the poet’s insistence on its importance has gone unregarded.  The immediate context of the poem provides an explanation for this insistence, and yet one might still wish that Roberts had taken the matter up directly.  To do so, however, would have required him to define, clearly and perhaps for the first time, some of what we now recognize as principal Canadian intellectual issues.

     An elegiac sense of the losses inflicted by time is not, of course, an exclusively Canadian feeling.  But there is an importantly Canadian attitude to the process of change which reflects the essentially conservative nature of the country in the nineteenth century.  This attitude is not widely shared today, when change is more often equated with progress than with loss.  So Elizabeth Waterston can say, of the conclusion of “Tantramar Revisited”: This is an ominous note of escapism in a young poet, still only twenty-five, in a young country very much in the process of change.”28  This criticism would have puzzled Roberts, and many of his contemporaries; the glorification of change was much more characteristic of the United States than of Canada, and Emerson was its most influential spokesman:

. . . let me remind the reader I am only an experimenter.  Do not set the least value on what I do, or the least discredit on what I do not, as if I pretened to settle anything as true or false.  I unsettle all things.  No facts are to me sacred; none are profane; I simply experiment, an endless seeker, with no past at my back.29

This leads us to Whitman, but not to Roberts, who argued (as part of the tradition so fully documented by Carl Berger in The Sense of Power) for a balancing of progressive and conservative impulses:

May we not accept liberalism with its enthusiastic energy unbated, while tempering its rashness with some of the enlightened conservatism to which study has taught that every blind extremist creates his own Nemesis, in the reaction which will overwhelm his efforts?  Reform is of doubtful desirability from the hands of narrow demagogues.30

Roberts’ “hesitancy” at the conclusion of his poem, his reluctance “to affirm the immediate moment,” is partly explainable then in terms of his Canadian circumstances.  His intellectual tradition allows him to accept change as process (the diurnal or seasonal cycles) but not necessarily as progress.  He may already have become uneasy about his ability to harmonize his own views with those of Goldwin Smith, a liberal and a continentalist.

     Roberts’ sense of dislocation would have been exacerbated, as well, by his strong sense of a regional identity.  From a centralist perspective his move to Toronto might possibly be viewed as beneficial; Arthur Lower speaks of “the sterile soil of New Brunswick which bred the Roberts family” and of the “half-light of provincialism” which blighted creative expression.31  An equally distinguished historian, A.G. Bailey, makes a somewhat different point about the cultural environment of his province:

[Bacon d’Avray] was one of a distinguished company that included some of the greatest teachers in the history of a province where teaching was practiced as an art.  Edwin Jacob, James Robb, George Roberts, William Brydone-Jack, Montgomery Campbell, George E. Foster, Henry Seabury Bridges, and their colleagues, were not the mediocrities that reviewers and writers of superficial biographies would have us believe.  Parkin testified otherwise when he wrote to Loring Bailey years later that in all his wanderings he had “seldom found social surroundings and intellectual influences more helpful and inspiring.”   “There was,” he said, “an old-fashioned courtesy and dignity — a real interest in things of the mind and spirit” in the University circle of Fredericton in the days of his youth.  Far from being a curious and inexplicable growth from a sterile soil, the poetry of Fredericton represented the flowering of a tradition that had been four generations in the making on the banks of the St. John; and behind that across the divide of the Revolution, lay the colonial centuries.32

In leaving Fredericton for Toronto, Roberts would not be exchanging rusticity for cosmopolitanism, and his hesitancy is, once again, understandable if we take his personal circumstances more fully into account than previous commentators on the poem have done.  That increasingly centralist view of Canada which dominated the period after Confederation, and achieved its greatest triumphs through Macdonald’s National Policy and the colonization of the west, had the effect as well of devaluing regional interests and identities.  To see the metropolis — Toronto — as the cultural centre became increasingly natural, and such an attitude was to be later reinforced by the influence of those American writers who led the “revolt from the village” in the period 1885-1925.  But Roberts’ “vantage-ground” offered him a different perspective, an intensely personal one, from which the loss of immediate contact with a regional culture could be seen as dislocation rather than progress.

     If we must say, then, that “Tantramar Revisited” falters at its conclusion, we should say so sympathetically.  Roberts was confronted by two problems which have, since 1883, continued to confront Canadians — and Canadian writers: how to be Canadian — to be North American without being continentalist; how to be national — to be at home in a nation without denying the value of regional cultures.  Roberts did not resolve his difficulties, and he had no literary models to which he might have turned for guidance.   Yet his admission of doubt is more admirable than the dismissive insouciance he would later display in “The Poet is Bidden to Manhattan Island”:

Dear Poet, quit your shady lanes
       And come where more than lanes are shady.
Leave Phyllis to the rustic swains
       And sing some Knickerbocker lady.
O hither haste, and here devise
       Divine ballades before unuttered.
Your poet’s eyes must recognize
       The side on which your bread is buttered!

Roberts was not the last Canadian writer to see such a move, whether physically or in spirit, as the best way of avoiding the problems of being a Canadian.  But “Tantramar Revisited” was one of the first Canadian poems to set those problems seriously before us, and it deserves its place in our literary tradition.


  1. William Strong, “Charles G.D. Roberts’ ‘The Tantramar Revisited’,” Canadian Poetry: Studies, Documents, Reviews, 3 (Fall Winter 1978), 26-37.  Page references hereafter cited in text.[back]

  2. Roberts and the Influence of his Time (Toronto: William Briggs,1905), p. 31.[back]

  3. Cappon, pp. 31,34.[back]

  4. Cappon, pp. 34-35.[back]

  5. Matters are not helped when, in offering us a line said to show Roberts’ metrical skill (the sixth line of the poem), he manages to misquote it: “Busy with spirit and [sic] flesh, all I have most [sic] adored”, Cappon, p. 35.[back]

  6. Sir Charles G.D. Roberts: A Biography (Toronto: Ryerson,1943), p. 48.[back]

  7. “Sir Charles G.D. Roberts,” in Ten Canadian Poets (Toronto: Ryerson,1958), pp. 47-48.[back]

  8. “Sir Charles G.D. Roberts,” in Essays in Canadian Criticism 1938-68 (Toronto: Ryerson 1969), p. 193; first delivered as a lecture at Carleton University, February 1961.[back]

  9. “Portraits of the Landscape as Poet: Canadian Nature as Aesthetic Symbol in Three Confederation Writers,” in Lorraine McMullen, ed., Twentieth-Century Essays on Confederation Literature (Ottawa: Tecumseh Press,1976), p. 148; originally published in the Journal of Canadian Studies, 6 (February 1971), 27-32.[back]

  10. Charles G.D. Roberts (Toronto: Copp Clark,1969), p. 48.[back]

  11. “Charles G.D. Roberts: Father of Canadian Poetry,” in Canadian Literature: Surrender or Revolution (Toronto: Steel Rail Educational Publishing, 1978), p. 51; an earlier version ofthis essay appeared in the Journal of Canadian Fiction, 1 (Winter 1972),47-56.[back]

  12. It should be noted that the “classical” (or quantitative) pentameter line did not consist of five dactylic feet, but was rather a hexameter line from which the short syllables of the third and sixth feet were dropped.[back]

  13. T. S.  Eliot noted that in Swinburne we usually find “emotion reinforced, not by intensification, but by expansion” (“Swinburne as Poet,” in The Sacred Wood [London: Methuen, 1920], p. 147), and Yvor Winters, after quoting part of a “characteristic” poem by Swinburne, commented that the style “is consistently controlled; it nowhere lapses from one mode to another.  It is resonant and pretentious” (Forms of Discovery [Denver: Alan Swallow,1967], p. 186).[back]

  14. Ten Canadian Poets, p. 48.[back]

  15. Essays in Canadian Criticism, pp. 192-193.[back]

  16. Strongs further suggestion — that we may find two other fourteen-line sections in the poem — seems to me less acceptable.  Seeing the poem in this way would lead us to disregard some important aspects of the five-part structure: the way in which the internal movement of the second section is emphasized by its being exactly half the length of the first, and the way in which the fourth and fifth sections constitute not a “whole” sonnet but a “broken” one (appropriate to the speaker’s hesitancy at the conclusion).[back]

  17. Charles G.D. Roberts, pp. 49,50.[back]

  18. Butterfly on Rock (Univ. of Toronto Press, 1970), p. 91.[back]

  19. “Portraits of the Landscape as Poet,” p. 148.[back]

  20. >Charles G.D. Roberts, p. 50.[back]

  21. Butterfly on Rock p. 91.[back]

  22. “Charles G.D. Roberts: Father of Canadian Poetry,” p. 50.[back]

  23. Selected Poetry and Critical Prose, p. 277.[back]

  24. Ibid., pp. 277-278.[back]

  25. “Portraits of the Landscape as Poet,” p. 149.[back]

  26. I say this despite the possibility that some readers will neglect Professor Jones’ point because he has, in paraphrasing part of the poem, referred to the hay that “waves golden on the hillsides,” notwithstanding Roberts’ own references “at this season” to “the fresh stowed hay in upland barns” and “yon cluster of haystacks.”[back]

  27. Selected Poetry and Critical Prose, pp. 244-245.[back]

  28. Survey: A Short History of Canadian Literature (Toronto: Methuen, 1973), p. 84.[back]

  29. Ralph Waldo Emerson, “Circles,” in Essays and Journals, selected and with an introduction by Lewis Mumford (New York: Doubleday, 1968), p. 220; originally published in Essays: First Series (1841).[back]

  30. Selected Poetry and Critical Prose, p. 247.[back]

  31. Colony to Nation (Toronto: Longmans,1946), p. 414.[back]

  32. “Creative Moments in the Culture of the Maritime Provinces,” in Culture and Nationality (Toronto: McClelland and Stewart, 1972), pp. 55-56.[back]