“It Seems Familiar but I Can’t Quite Remember”: Amnesia and the Dislocation of History and Gender in Random Harvest (1942)

“Whose stories do these films really tell?”

Voice, point-of-view, and the stories films tell are inextricably tied to the critical and academic narratives that have been told about them. Nowhere is this more evident than in the now-large body of work surrounding the classical Hollywood woman’s film, work that probably began with Molly Haskell’s From Reverence to Rape (1973), reached its apex with Mary Ann Doane’s The Desire to Desire (1987), and persists today in current feminist and post-feminist considerations of classical and contemporary “chick flicks” and newer media forms. Issues of speech, voice, subjectivity, narration, and image haunt both the woman’s film and the many studies of the genre, and they will no doubt continue to do so for as long as these films attract audiences, however different contemporary audiences may be from those who originally watched them. Rather than revisit these subjects as discrete issues, however, I am interested in exploring the ways in which they intersect with history and what I’m calling a dislocation of gender that takes place in some woman’s films that address history. By “history” I refer to both the moment of a film’s production as well as a narrative time-frame and setting overtly represented as events of historical record.

As I have noted elsewhere (2009), it is common to assert that Hollywood film generally represents historical events in terms of the story of its impact upon individuals rather than upon larger social, political, cultural, or economic groups. However, much remains to be said about the discursive relationship that the woman’s film constructs among history, narrative, gender, and subjectivity. If, as David Bordwell, Maureen Turim, and myriad film historians have pointed out in the past, classical Hollywood films represent history neither accurately nor in sufficient complexity, what exactly do they do with history? Specifically, for the purposes of my investigation here, how do some Hollywood woman’s films combine issues of history and gender in ways that are narratively meaningful, if rarely historically accurate or ideologically progressive? Moroever, how do answers to these questions further our understanding of what is at stake in the classical woman’s film and in existing critical studies of them? Because it directly takes up the first three questions, I will discuss Mervyn LeRoy’s 1942 film Random Harvest at some length, though this single film is simply an explicit example of the dislocation of gender and history that occurs in a number of other woman’s films as well.

(Re)Defining the Woman’s Film

The “woman’s film” is a somewhat ambiguous term. Recognized as a production category by major Hollywood studios in the 1930s and 1940s, it has assumed a number of related but different meanings within theoretical and critical debates in film studies over the past three and a half decades. On the one hand, historian Tino Balio, acknowledging the work of feminists Molly Haskell and Lea Jacobs in particular, characterized the woman’s film of the 1930s in particular as a “cycle,” one that typically refers to a type of motion picture that revolves around an adult female protagonist and is designed to appeal mostly to a female audience . . . Woman’s film is a term of convenience to describe a range of pictures commonly referred to as fallen-woman films, romantic drama, Cinderella romances, and gold-digger or working-girl stories. The titles of such pictures (e.g., Faithless, Shopworn, Stella Dallas) are often taken from the names of their heroines or make some reference to women’s conditions. The conflicts in the pictures involve interpersonal relationships that present the heroine with dilemmas the resolutions of which usually entail loss.

Balio’s definition is accurate and a useful departure point, but it is interesting to me now because of the total absence of history — as event or representation — in Balio’s reference to “women’s conditions.” In this characterization, woman’s films would seem to have nothing to do with history, either as an issue that a movie-going public would consider relevant to women or at the level of diegesis.

On the other hand, in her canonical work on the woman’s films of the 1940s, Mary Ann Doane considered the films of that decade a full-fledged genre rather than a cycle. Unlike Balio and emblematic of the preoccupations of second-wave feminist film scholars, Doane was less interested in plot patterns (though certainly plot is a consideration of her work) than she was in subject matter and the specifically cinematic enunciative strategies that position the spectating public, presumed by the American film industry of the period to be primarily female:

The label “woman’s film” refers to a genre of Hollywood films produced from the silent era through the 1950s and early ’60s but most heavily concentrated and most popular in the 1930s and ’40s. The films deal with a female protagonist and often appear to allow her significant access to point of view structures and the enunciative level of the filmic discourse. They treat problems defined as “female” (problems revolving around domestic life, the family, children, self-sacrifice, and the relationship between women and production vs. that between women and reproduction), and, most crucially, are directed toward a female audience. (3)

Eloquently argued, Doane’s study examined the woman’s film as “a privileged site for the analysis of the given terms of female spectatorship and the inscription of subjectivity.” She was concerned with analyzing the woman’s film primarily in terms of its address to the female spectator and isolated the 1940s as a period in which World War II caused the Hollywood film industry to anticipate a largely female audience and to generate woman’s films that addressed the female spectator more directly and specifically than other genres and other woman’s films in other periods. Doane’s work remains the definitive study of the woman’s film of the 1940s, but it does not specifically address the intersection of gender and representations of history within the woman’s film, nor what, if anything, might be at stake in them.

Contested Tears: Whose Voices Matter?

At the foundation of my ensuing remarks is the still-unresolved (and perhaps definitively unresolvable) question that has haunted the woman’s film and the many who have watched and studied them for decades now: whose stories do these films really tell? Whose voices do we hear, whose voices matter, and — in the case of woman’s films that explicitly engage with or evoke history — whose stories are told and what, if any, relationship do these films construct between history and gender? These same questions also reside at the heart of the “history” constructed about the woman’s film in and through academic discourse of the past 35 years as well. That discourse has shaped our perceptions of generic parameters, œuvres, and issues deemed worthy of scholarly debate, and it also has participated in wider conventions about critical and authorial legitimacy within academic film and media studies itself.

To illustrate my point, let me briefly invoke the mid-1990s work and reception of American philosopher Stanley Cavell in Contesting Tears: The Hollywood Melodrama of the Unknown Woman. Constructing his argument in both implicit and explicit dialogue with his earlier work on Hollywood romantic comedies (1976), Cavell maintained that four films — Stella Dallas (King Vidor, 1937), Now, Voyager (Irving Rapper, 1942), Gaslight (George Cukor, 1944), and Letter from an Unknown Woman (Max Ophuls, 1948) — constitute a genre he called the unknown woman film, and that these films “recount interacting versions of a story, a story of myth, that seems to present itself as a woman’s search for a story, or of the right to tell her story” (3). An unsurprising assertion to those familiar with the extensive work in film studies on melodrama and the woman’s film that was done prior to Cavell’s book, Contesting Tears considers the woman’s film not in relation to issues of gender, feminism, or its related critical discourses, but rather to Cavell’s own earlier study of what he has termed “comedies of remarriage.” Thus, Cavell positioned himself outside the notable and highly visible work that sought variously to identify melodrama as a trans-generic mode of representation (e.g., Brooks, Williams), deconstruct a gendered alignment of subjectivity within the woman’s film within a specific decade (e.g., Doane), or locate issues of mode, genre, gender, and representation within a particular social and historical moment (e.g., Gledhill, Byars) — to cite merely a few instances of that work. As a result, in responses to earlier drafts and presentations of his own study as well as to the final book itself, Cavell found himself criticized for ignoring the work of feminists and film scholars (see, in particular, responses and reviews by Modleski 1990 and Kaplan 1998). Asserting “I have no standing, and no motive, from which to attempt to place these different emphases nor to seek out others,” Cavell argued for the legitimacy of his own interest in various relationships among cinema, philosophical skepticism, tragedy, melodrama, gender, and psychoanalysis, choosing to stay largely within the terms of philosophical debate most familiar to him. Sketching a timeline for the evolution of his interests (to demonstrate their autonomy and integrity), he simultaneously acknowledged that the work of feminist film scholars existed, declared his independence from that work, and curiously remarked upon his “sense for a long time of intellectual isolation” (CT, 199) as his interests developed.

Feminists might (and did) argue that Cavell’s was a self-imposed isolation, but E. Ann Kaplan’s assessment of his book in Film Quarterly is interesting to me today for two reasons. First, Kaplan engaged in a conscientious effort to assess Cavell’s work on the woman’s film and acknowledged her own investment in the controversy surrounding it. Second, and perhaps of more interest now, she identified what she called “anxieties of time and gender” (as well as discipline) at work in Cavell’s book (78). Though she did not discuss them as such in her review, time and gender are precisely the well-known twin anxieties that haunt melodrama and the woman’s film specifically, with the melodramatic “too late” generally tied to an experience of women’s waiting (for love, for recognition) that may never come or may come past the point to make any difference (Doane, 1987, Modleski 1988). Read today, Kaplan’s observation functions almost as a gloss not only on the genre of the woman’s film but on a curious element of the debate among Cavell and his feminist critics: it played out as a kind of discursive melodrama in academic circles, with Cavell himself taking up the traditional position of the woman as the one who waits (for love, for recognition from his academic peers). Cavell’s anxiety of time, Kaplan said, emerged partly as his “need to lay claim to the authority that comes from doing research for so long. Perhaps there is anxiety about running out of time [emphasis mine] to complete his projects” (78). As for the anxiety of gender, Kaplan pointed to its expression in Cavell’s choice not to engage with long-standing feminist debates about the woman’s film that both preceded and were occurring simultaneously with the development of his work:

[H]is excuse for not paying attention to the feminist melodrama work because not ‘specifically invited to’ still strikes me as disingenuous. Cavell evidently only engages with research that is in line with his own thoughts: he sees no need to take other points of view into account. (emphasis in original, 78)

Dramas of invitation and rejection, acknowledgment and dismissal, play out in Cavell’s own words as he levels a somewhat inverted claim at his critics, recasting it in somewhat labored but wounded prose:

[E]ven I, for all my overlaps yet asynchronies with the interests of my culture, have had to recognize that the expression of intellectual indebtedness or helpfulness is no longer dischargeable on exactly intellectual grounds. No doubt it never was. But it is as if a current preoccupation with an [anti] metaphysics of citationality and of authorship have come to mask a politics of who is citable by whom and who not. (CT, 199)

If Kaplan’s review unselfconsciously evokes tropes of the trajectory of a woman’s film, Cavell’s own words might be said to correspond to scenarios more similar to the 1950s family melodrama of authority and privilege about which Thomas Elsaesser wrote so compellingly in “Tales of Sound and Fury” (1972). A palpable sense of resentment and disappointment seems to permeate Cavell’s comments here: they are a grudging, back-handed admission that, although he may not have been influenced by others outside his discipline working on related material, such materials existed. They are an expression of irritation that his work should “have to” allude to or be assessed in any kind of relation to that material. They also give voice to his disappointment that, just as he may not have considered the work of feminist and other film scholars directly relevant to his project, those same people, in turn, have chosen not to engage as fully with his work as he might have wished.

I refer to the debate around Cavell’s work on the woman’s film not to centralize the work or the debate unduly but rather to observe that, taken together, they dramatize issues that subtend critical discussion within any academic field (whose voices “count” and are acknowledged as relevant or authoritative and why?). And, perhaps not coincidentally, issues of authority, voice, and interpretation are also matters that many woman’s films emphasize through plot and assorted enunciative strategies, both visual and aural. My focus for the remainder of this piece will shift to discuss how history — as event and representation — intersects with issues of gender in the woman’s film, specifically in terms of the 1942 MGM film Random Harvest. Whose voices and stories are most clearly articulated? How are personal and public histories linked? At what levels and in what ways are they given narrative expression? Ultimately I argue that the film’s construction of story, history, and identity productively dislocates (rather than underpins) binary oppositions between history and story. I argue, too, that through the narrativization of two World Wars in personal terms, gender itself is not firmly anchored to body and image (as it has been so often thought to be in the woman’s film) but is itself dislocated across the complicated terrain of the narrative.

Dislocation of History

At first glance, Random Harvest would seem to be less of a woman’s film, per se, as defined by Balio or Doane, and more of a vehicle for the well-established male star and romantic lead Ronald Colman, who receives first billing, and only secondarily for Greer Garson, who had become immediately successful in her first film (and another adaptation of a James Hilton novel) Goodbye Mr. Chips two years earlier. It focuses on the loss of memory of a male character, whom the film follows as he recalls one life only to forget another. The film introduces an anonymous World War I veteran and amnesiac (“John Smith,” played by Colman) forming new ties and marrying musical hall performer Paula (Garson) in the absence of any remembered previous existence, then adapting to his former unrecalled identity as wealthy, unmarried businessman Charles Rainier once his memory is unexpectedly recovered (and his short life as John Smith forgotten). A significant amount of screen time elapses before Paula appears in the film, and this delay is replicated later when Smithy remembers his life as Ranier. Paula reappears in the film later as Ranier’s quietly devoted secretary Margaret, who has clearly followed Smithy/Charles through the years and waits for him to remember her. Initially, a male voice-over accompanies the opening images of the film and is supplied by the author of Random Harvest himself, James Hilton (who also provided the voice-overs for Madame Curie, another Mervyn LeRoy/Greer Garson teaming [1944], as well for an adaptation of another of Hilton’s books So Well Remembered [Edward Dmytryk, 1947]). Pinning narrational and authorial voice-over to male star, the film at least initially underscores Colman/Smithy/Ranier’s subjectivity and experience rather than Garson/Paula/Margaret’s.

However, the film’s incipient narrative focus on Smithy and the question of his former identity and its later shift to Paula do more than dislocate issues of gender and narrative across multiple sites, as I will discuss. In its invocation of World War I, it also explicitly structures history in relation to gender in ways that other woman’s films do not. In her work from the late 1980s on flashbacks in film, Maureen Turim noted that a number of films from World War II use the rhetorical device of the flashback both to subjectivize history and to construct a relationship between two different historical periods:

The Second World War marks the appearance of a number of films that not only use flashbacks to subjectivize history . . . but also to compare directly through their flashback construction two different historical periods. Taking as their present moment the War, either on the home front or active duty overseas, these films compare experience of that War with World War I . . . These films render historical thought as the subjective experience of individuals by creating both the present and the past which inserts itself in the present as a series of highly personal experiences. The major events of history are represented as they affect the course of daily life, particularly as they affect the course of romance, marriage, or motherhood. (122-123)

Turim’s remarks here apply equally to Random Harvest, though it is not a film in which flashbacks figure. Despite the absence of flashbacks, however, its almost obsessive fascination with an unrecalled personal past weaves a relationship among not just two but three time frames: 1) the historical and narrative past of World War I and the context for Charles’ amnesia; 2) the “present” defined in the film’s story (a briefly glimpsed invitation within the diegesis identifies that present as 1935); 3) the moment of the film’s production and release, insofar as the fashions in the film (an aspect of mise-en-scène) evoke the year 1942 far more clearly than they do 1935. While an overt narrative connection between the first and second World Wars is not part of any flashback structure in Random Harvest, an important link between the two wars converges in the film’s original 1942 viewing audience. That is, Random Harvest converts the experiences of the First World War into a tale about loss and recovery that are experienced by both Smithy/Charles and Paula/Margaret. The invocation of World War I takes place against a contemporary audience’s experience of 1942 (a bleak year for the Allies during the Second World War) and renders the latter war meaningful in light of the former. The characters’ personal experiences of renunciation and suffering occasioned by the First World War would have been inevitably understood by at least some contemporary audience members in terms of their own remembrance of that earlier war, as well as their present-day experience of deprivation and loss. This is particularly true in 1942, a year in which public discourse in the United States focused the need for large-scale sacrifice and patience on the home front. In 1942 the War Advertising Council, in conjunction with different government agencies, launched its several campaigns in support of the United States war effort: “War Bonds” (1942-1945, sponsored by the U.S. Department of Treasury [War Finance Agency]), “Security of War Information — Loose Lips Sink Ships” (1942-1945, sponsored by the Office of War Information, the U.S. Army, the U.S. Navy and the Federal Bureau of Investigation), and “Women in War Jobs — Rosie the Riveter” (1942-1945, sponsored by the Office of War Information and the War Manpower Commission). In keeping with such ideological messages about the necessity for thrift, silence, forbearance, and hard work, audiences watched as Paula/Margaret waits and watches for Smithy/Charles, and Smithy/Charles does his own share of waiting as well (even if what he waits upon is his own fractured memory to coalesce). That waiting assumes both intra- and extra-textual levels of gendered meaning in relation to the history that the film depicts and is set against. Narrational point-of-view structures and dynamics, gender, and history are dislocated across multiple axes, finding their points of intersection not so much in a monolithic set of optics or aural cues but in the diverse sense-making capabilities of the audience(s) for whom the narrative unfolds.

Dislocation of Gender

“Sometimes I have a feeling of being — if it isn’t too absurd to say such a thing — of being half somebody else.” — Charles Rainier, from James Hilton, Random Harvest(italics in original)

The epigraph with which I begin this section is spoken in the novel by Charles Rainier as he struggles with the wisp of a memory from his short, forgotten life as John Smith, or “Smithy,” as Paula calls him. The words point eloquently to the way in which the historical woman’s film based on that same novel questions the relationship among gender, history, and identity. It is not simply that historical woman’s films subjectivize history, as Turim and Bordwell have pointed out (although that is something that they undeniably do). Rather, it is that gender itself is often somewhat disembodied and dislocated in the blurring of lines between male and female protagonists’ experience of a historical (if personalized) past, and this dislocation is articulated not only at the level of plot but also as part of the film’s visual and aural logic. Elsewhere I have suggested how this migration and dispersal occurs in relation to both other historical woman’s films (2009) and occasionally to those woman’s films that do not overtly invoke history (2005). My subsequent discussion here will suggest some of the ways in which a similar process occurs in Random Harvest.

Although the film begins in a way that suggests the story we are about to watch is Smithy/Charles’ own, as I have observed, it is not that simple. From the start, Smithy and then Charles are characterized as men whose capabilities and powers are emphatically diminished and restricted, while Paula is constructed as an additional source of strong narrative interest, action, and organizing point-of-view. The film opens in Melbridge County Asylum in 1918: a succession of interior scenes introduces the hospital, its presiding psychiatrist Jonathan Benet (Philip Dorn), and Smithy, a nameless survivor from the battlegrounds of Arras in 1917. Paternalistic in the extreme, Benet refers to all his patients as “my boy,” regardless of their age (and in the case of Colman, born 10 years before Dorn, the epithet hangs in the air almost ludicrously. An anonymous contemporary reviewer observed that “Colman gives a fine performance but is not quite the romantic type that he was years ago. In fact, he looks older than he should have been for film expediency” [Variety, January 1, 1942]). Nevertheless, the psychiatrist’s repeated references and the extreme difficulty with which Colman’s character speaks, combined with his lack of memory, emphasize his passivity and powerlessness, as does the mise-en-scène. In the second shot in which he appears, Colman stands at a distance from the camera, framed by a window in the background behind him and by the two figures of the psychiatrist and his assistant in the immediate foreground: the effect is to diminish and convey a graphic sense of entrapment associated with Colman’s character.

Smithy’s and Charles’ disempowered state persists as the film continues, initially emphasized at the visual level, as I have described, and then at the level of the soundtrack. Ultimately Paula is introduced as alternating foil to and mirror of Smithy and then Charles in a progressive dislocation of a single locus of gendered identity or desire. The anonymous male voice-over might first associate the film with a strictly masculine point-of-view, but both the anonymity of the narrator and the voice-over’s brevity undermine its significance as well as its authority. Indeed, after it disappears following the opening shots of the film, that voice-over is metonymically replaced by Greer Garson’s voice in the sequence in which she and Colman’s character meet for the first time. The combined effect of plot events, visual presentation, and aural detail work together to further dislocate and disembody a stable sense of identity or authority (“Sometimes I have a feeling of being — if it isn’t too absurd to say such a thing — of being half somebody else“). When Smithy and Paula first meet, the camera is on Colman, but a soft yet clear woman’s voice is heard offscreen — “You are from the asylum, aren’t you?” — and the camera pans left, away from Smithy to reveal Paula, watching him with interest and evident sympathy. A few minutes later, in another location, a pub keeper pours Colman’s character a complimentary drink, observing in hearty tones that “That’ll bring anybody back from the grave!” Not only is Paula’s fluid, closely miked speech infinitely more articulate than that of Smithy (or the narrator Hilton, for that matter), but also her voice, momentarily offscreen, has the power to draw the camera away from Smithy’s face and to her own. The shot/camera movement functions as a visual emblem for the dynamics of the narrative as a whole: this may first seem to be a story belonging to Ronald Colman’s character, but Greer Garson is revealed as a twin source of narrative interest and movement. (Indeed, the same contemporary reviewer I  alluded to a moment ago also notes “Greer Garson . . . is an important mainstay of the picture. Essaying a highly sympathetic role, she overshadows Colman.” [Variety, January 1, 1942]). Garson’s presence in the film, as well as its preoccupation with love, romance, self-sacrifice, and suffering, marks it as both a woman’s film and a vehicle for Colman. However, its evocation of a dislocated subjectivity and a nomadic desire eludes the simple binarisms so often reductively associated with the woman’s film as a genre.

Such blurring of neat boundaries and divisions, I believe, occurs more frequently in classical cinema than was originally theorized in the 1970s and ’80s, but it does so across a range of discursive levels throughout the course of a narrative, rather than within a single discursive realm, such as the purely visual or aural. Film narrative was theorized by ideological and feminist critics in the 1970s and 1980s largely as a relay of looks within and at film at the cost of other signifying discourses. However, when one examines the visual register in conjunction with sound and issues of narrative dynamics, the deployment of gender and desire within and across narrative becomes infinitely more complex. (See, for example, Britta Sjogren’s rereading of Kaja Silverman and Mary Ann Doane, 2006).

As Smithy/Charles remains almost wholly incapacitated by amnesia throughout the film, uncertain of his identity in any single moment, of his present as well as his past, Paula takes on an active role in the first and third parts of the film. His niece Kitty, her surrogate, takes over that role during the middle portion of the narrative when Paula is absent. Paula notes Smithy’s obvious helplessness when they first meet (“Will you be all right? . . . What am I to do with you?”), gives him the name “Smithy” when he reveals he hasn’t got one, and all but proposes to him later, telling him after he declares his love for her that, “I’ve run after you from the beginning. I’ve never let you out of my sight since I first saw you in that little shop.” Established as the more active of the two, Paula even has to tell Smithy what to do after she accepts his marriage proposal: “Smithy, do I always have to take the initiative? You’re supposed to kiss me!”

Even after Smithy has returned to his life as Charles Rainier, a man who occupies a position of considerable social and economic power, he remains curiously inactive; instead, it is the desires of those around him — specifically, of Kitty for him — that both constitute the plot and activate the narrative movement. As incapacitated by a past he cannot remember as Smithy was, and without the now-forgotten Paula to guide him, Charles forms an attachment to Kitty. However, that attachment is as passive as his desire for Paula was. Instead, in those sections of the film from which Paula is absent, Kitty replaces her as the narrative’s driving force, the textual energy of the film handed off in almost baton-like fashion from woman to woman as desire itself is relayed across the narrative terrain of the film. Like Paula before her, Kitty takes the initiative in her relationship with Charles in scenes that recall Paula’s actions toward Smithy.

It is important to acknowledge here that Mary Ann Doane has taken note of similar manifestations of passivity in male protagonists within the woman’s film and read them as evidence of a process of feminization:

[T]he genre does seem to require that the male character undergo a process of feminization by his presence within a love story . . . . The feminization of the male . . . facilitates the female spectator’s divided identification with both the man and the woman — the man desiring the woman, the woman desired by the man . . . Perhaps this is why, in the love story, it is frequently difficult to tell whether the man or the woman is the protagonist . . . The double identification of the female spectator enables a confusion whereby the woman’s story is represented as the memory or vision of the man. (116-117)

For Doane, identification and gender in the woman’s film were definitively (though not always straightforwardly) linked in such a way that even the (imputed) female spectator’s gaze at a film is ultimately recouped by the organizing gaze of the cinematic apparatus which she identified as male. However, I view processes of narrative identification in film as less firmly defined by the look of the camera, and the look of the camera less definitively tied to gender. Male and female spectators — imputed and actual — identify variously and contradictorily with different characters and with the sheer dynamics of movement across the unfolding of story and space. What Doane reads as narrative difficulty or confusion engendered by a film’s effort to entertain a female gaze, I read as evidence of a less pessimistic, more fluid articulation of gender in the woman’s film. On one level, both Paula/Margaret and Kitty wait for Smithy/Charles to see them, his recognition determining their possibility for happiness in a paradigmatic structure of power and looking relations that is often theorized as foundational to visual pleasure in western patriarchal representation. On another level, however, Charles’ ability to see clearly is so compromised that the visual register of the film cannot be taken as its only meaningful discursive realm. Thus, I would also argue that neither Doane nor I is definitively “right” or “wrong” in our respective readings and that every academic is as much a product of his or her own moment of intellectual thought and influence as the objects they study.

However, ultimately I am arguing for a more nuanced understanding of the relationship among gender, history, and subjectivity in 1940s woman’s films than has been previously acknowledged, and for the concept of a multivalent organizing subjectivity that is occasionally positionless and bodiless, one that traverses the divide between masculine and feminine at the level of plot as well as cinematic form. In an expository montage that simultaneously details Kitty’s maturation from adolescent schoolgirl to grown woman and Charles’ rise in the business world, Kitty writes to Charles in what is obviously a stream of letters over time: (“Dear Uncle Charles . . . Dear Charles . . . My dear Charles . . . “). The initial images of her writing to Charles, speaking her letter aloud, dissolve first to a photograph of her as a graduate (“I’m sending you my photograph in cap and gown. Will you put me on the desk in your study? And please, look at me sometimes”), then as a young woman in a formal gown (“Just to remind you that I’m growing up. I’ve lots of beaux. I do hope you’re jealous”) with Kitty’s voice-over as accompaniment. It would seem that Charles has complied with her request to place her photograph on his desk, as we see it on a desk, near a window, through which the changing seasons can be glimpsed. But the montage itself conveys a curious blending of perspectives that are simultaneously Charles’ as well as Kitty’s. The image itself suggests, perhaps, that the gaze of the camera is aligned with an offscreen Charles who is looking at the photograph: Kitty wants desperately to be seen by Charles (as Paula/Margaret does) and, in the photograph, stares into the lens of the camera. On the other hand, there is no establishing or reverse shot of Charles looking at Kitty, and in its absence and given the fact of Kitty’s voice-over, the scene may also be read as an evocation of Kitty’s fantasy of her photograph on the desk in Charles’ study, of her wish — expressed in her voice-over — fulfilled. The sequence is interesting precisely for its ambiguity: subjectivity hovers across the gender divide, neither Charles’ nor Kitty’s, but rather both at once.

It is precisely these moments of ambiguity, far from exceptional in the 1940s woman’s film but not often commented upon heretofore, that complicate the warp and woof of the narrative fabric of gendered desire (McKee, 2005). In an article devoted to Joan Leslie’s novel The Ghost and Mrs. Muir on which the film of the same title is based, Margaret D. Steitz explained the rise of the post-war ghost in terms of the experience of World War II by actual women:

Surely many of the women who emerged from the nightmare of the Second World War, especially those who had also lived through the First, felt themselves . . . caught in a liminal world, a place inhabited by specters [emphasis mine]. For such women, gothic conventions and metaphors provided the nearest equivalent to a language with which to describe that mental state and to render its strangeness more familiar. But it was gothic comedy or mock-gothic narrative that could offer the safest framework through which to acknowledge this condition of being “haunted” and yet to allow the terrors attached to that recognition. (20)

Neither gothic nor mock-gothic, Random Harvest nevertheless evokes the metaphor of the ghost in relation to Charles and Margaret equally. As Smithy/Charles does not have access to his past — to his own history which is inextricably bound up in the first World War and his relationship with Paula — he is, essentially, a shade, condemned to wander passively in a perpetual half-life. In addition to the pub keeper’s and Kitty’s allusions to Charles as someone who has been brought back from the dead, Paula has Smithy declared legally dead upon learning of his plans to marry Kitty. However, Paula herself has her own preoccupations with the past, and in addition to being more active than Charles, she is also his feminine twin: Charles asks Margaret to marry him for the assets she will bring to him as hostess and organizer when his career takes a political turn, telling her, “You and I are in the same boat: we’re ghost-ridden prisoners of our pasts.” She assents, but eventually the strain of marrying the same man twice — first to be forgotten, then to go unrecognized — tells on her. One evening she observes to Charles, with an edge to her voice, that “The best of you — the capacity for loving, the joy in living — is buried in a little space of time that you’ve forgotten.” Provoked by Charles’ comment that she has buried her own heart, she pushes him: “You may have come so near [the woman you’ve forgotten], might have brushed by her in the street . . . You might even have met her, Charles — met her and not known her. She might be someone you know. Charles — she might even be me!”

Conclusion

What haunts, finally, after Random Harvest‘s final fade to black, at the moment in which Charles has at long last recognized Margaret as Paula and embraced her, is the way in which personal memory and historical event have been inextricably bound up and rendered meaningful in relation to each other for the film’s original audience and, more faintly, for its more recent audiences as well. Charles’ words in Hilton’s novel — “Sometimes I have a feeling of being . . . half somebody else” — work as a kind of palimpsest for the film. They evoke not only Charles’ own alternate identity as Smithy, brought into being by the traumatic experience of war, but also a much older story: Plato’s account of the original androgyne who was cleaved in two and had to search for its missing half. Romanticized by Western culture through the centuries as a lover’s search for an ideal soul mate, Plato’s tale points to Charles’ missing half as both himself (Smithy) and Paula. It also points rather eloquently to the ways in which gender and history are dislocated rather than immutably fixed to body and image, traversing the terrain of the love story from Plato to the 1940s woman’s film and beyond.

Works Cited

Anonymous. Variety, January 1, 1942

Balio, Tino. Grand Design: Hollywood as a Modern Business Enterprise 1930-1939. New York: Charles Scribner’s Sons, 1993.

Bordwell, David, Janet Staiger and Kristin Thompson, The Classical Hollywood Cinema: Film Style & Mode of Production to 1960. New York: Columbia University Press, 1985.

Brooks, Peter. The Melodramatic Imagination: Balzac, Henry James, Melodrama, and the Mode of Excess. New York: Columbia University Press, 1984. Originally published in 1976 by Yale University.

_____. Reading for the Plot: Design and Intention in Narrative. New York: Random House, 1984.

Byars, Jackie. All That Hollywood Allows: Re-reading Gender in 1950s Melodrama. Chapel Hill and London: The University of North Carolina Press, 1991.

Cavell, Stanley. Contesting Tears: The Hollywood Melodrama of the Unknown Woman. Chicago and London: University of Chicago Press, 1996.

_____. Pursuits of Happiness: The Hollywood Comedy of Remarriage. Cambrige, MA: Harvard University Press, [1976] 1981.

Doane, Mary Ann. The Desire to Desire: The Woman’s Film of the 1940s. Bloomington and Indianapolis, 1987.

Elsaesser, Thomas. “Tales of Sound and Fury: Observations on the Family Melodrama.”Monogram 4 (1972).

Gledhill, Christine, ed. Home Is Where the Heart Is: Studies in Melodrama and the Woman’s Film. London: BFI Publishing, 1987.

Harper, Sue. “Historical Pleasures: Gainsborough Costume Melodrama.” In Home Is Where the Heart Is: Studies in Melodrama and the Woman’s Film.

Haskell, Molly. From Reverence to Rape: The Treatment of Women in the Movies. New York: Holt, Rinehart and Winston, Inc, 1973.

Hilton, James. Random Harvest. Boston: Little, Brown & Company, 1941.

Jacobs, Lea. The Wages of Sin: Censorship and the Fallen Woman Film, 1928-1942. Madison: University of Wisconsin Press, 1991.

Kaplan, E. Ann. Untitled review. Contesting Tears: The Hollywood Melodrama of the Unknown Woman. Film Quarterly 52:1 (Autumn 1998), 77-81.

McKee, Alison L. “‘L’affaire Praslin’ and All This, and Heaven Too: Gender, Genre, and History in the 1940s Woman’s Film.” The Velvet Light Trap, Spring 1995.

_____. “Rewriting Desire from a Feminist Perspective: Classical Film Narrative and The Ghost and Mrs. Muir (1947).” The Bulletin of the Institute of Language and Culture. Meiji Gakuin University, Tokyo, Vol. 23, 2006 (Japanese translation).

_____. “She Had Eyes a Man Could Drown In: Narrative, Desire, and the Female Gaze in The French Lieutenant’s Woman.” Literature/Film Quarterly 20:2, 1992.

_____. “What’s Love Got to Do with It? History and Melodrama in the 1940s Woman’s Film.” Forthcoming in Film & History: An Interdisciplinary Journal 39:2 (Fall 2009).

Modleski, Tania. “Time and Desire in the Woman’s Film,” in Home is Where the Heart Is: Studies in Melodrama and the Woman’s Film, 335.

_____. “Reply to Cavell.” Critical Inquiry 16. Autumn 1990.

Mulvey, Laura. “Visual Pleasure and Narrative Cinema.” Screen 16:3 (Autumn 1975).

Silverman, Kaja. The Acoustic Mirror: The Female Voice in Psychoanalysis and Cinema. Bloomington and Indiana: Indiana University Press, 1988.

Sjogren, Britta. Into the Vortex: Female Voice and Paradox in Film. University of Illinois, 2006.

Margaret D, Steitz, “The Ghost and Mrs. Muir: Laughing with the Captain of the House,”Studies in the Novel 38:1 (Spring 1996).

Turim, Maureen. Flashbacks in Film: Memory & History. New York and London: Routledge, 1989.

Williams, Linda. “Film Bodies: Gender, Genre, and Excess.” Film Quarterly 44:4 (Summer 1991).