The AI Journal

Harriet Beecher Stowe kills Little Eva slowly, stretching the deathbed scene across three chapters and several thousand words of a novel that had already killed off one child. A modern reader who assumes the serial format demanded the padding is wrong: when Uncle Tom’s Cabin left the pages of the National Era for a single bound volume in 1852, the death stayed exactly as long as it had run in weekly instalments. Its length was not an accident of publication. It was the novel’s own claim on the reader’s attention, and for a century after, that claim was cited as the clearest evidence against the kind of fiction Stowe was writing: manufactured feeling standing in for the psychological complexity a serious novel was supposed to supply instead.

Ann Douglas’s The Feminization of American Culture (1977) gave that judgement its most influential historical form. Mid-century sentimental writing, on her account, was produced largely by women and clergy who had lost real cultural authority as American religion secularised and commerce hardened into a competitive, masculine public sphere; sentimentalism replaced theological seriousness with a culture of feeling that flattered its readers and asked nothing rigorous of them. This was not simply a complaint that individual sentimental novels were badly made. It was a diagnosis: the genre’s characteristic moves, its repetitions, its direct appeals, its manufactured deathbeds, were symptoms of a decline Douglas traced forward into a modern American taste for consoling banality. The thesis is powerful because it does not need any single sentimental novel to be a failure by its own standards. It only needs those standards to be the wrong ones to be applying, and it assumes without much argument that they are the right ones.

Jane Tompkins’s Sensational Designs (1985) answered from a different direction. Rather than defending sentimental fiction against the charge of bad craftsmanship, she rejected the standard doing the judging. Uncle Tom’s Cabin was not attempting the psychological interiority and ironic distance by which the American canon had learned to measure fiction; it was attempting what Tompkins called cultural work on readers who already shared its theological and political premises, using techniques calibrated to that readership rather than to a later one. Typology, direct address, and the extended deathbed as proof of a soul’s readiness were not failed attempts at realism. They were an argument, aimed at evangelical Protestants who needed to be moved from tolerating slavery in the abstract to opposing it in practice, and the novel that helped do that to hundreds of thousands of readers was succeeding at something, whatever name is given to the thing it was not.

Nina Baym’s Woman’s Fiction (1978) supplies the evidence that this was not particular to Stowe or peculiar to the antislavery novel. Baym read roughly a hundred and thirty novels by four dozen women published between 1820 and 1870 and found a stable underlying plot: a girl loses the financial and familial support she was raised to expect, and through a sequence of trials acquires the self-command to survive without either. Counted novel by novel, the repetition looks like a failure of imagination. Counted across the genre, it looks like a shared vocabulary, the same formal move deployed with the reliability a reader expects from a genre and rewards with continued purchase, doing a describable job for a describable audience rather than failing at a different job nobody in that audience had asked it to do.

Joanne Dobson’s 1997 essay in American Literature pushes the defence past the historicist claim that sentimental fiction did cultural work and into a claim about the sentence itself. Dobson catalogues the genre’s recurring figures, severed hands, ruptured domestic circles, watched deathbeds, and argues that their plainness of diction and reliance on physical metaphors of connection are not the absence of craft but craft aimed at a specific end: legibility. A sentimental novel that buried its emotional argument behind ambiguity would have failed on its own terms, whatever it might have gained on someone else’s. The charge that sentimental prose is unliterary, on this reading, mistakes one set of literary priorities, communicability over ambiguity, for the absence of any priorities at all.

None of this requires denying that most sentimental fiction is, by the standards its detractors apply, bad: repetitive within a novel as well as across a shelf of them, thin on interiority, content with a plot a practised reader could predict by page thirty. It requires only separating two questions that the charge of artlessness conflates: whether a text achieves the effects a different aesthetic tradition prizes, and whether it achieves the effects its own tradition prizes. Most of the evidence marshalled against sentimental fiction over the past century answers the first question while presenting itself as an answer to the second, and Tompkins, Baym and Dobson are together making the same point from three angles: measured against its own genre’s aims, the sentimental novel was frequently doing exactly what it was built to do.

The strongest objection to this defence does not dispute the craft. It accepts it and asks what the craft was for. Karen Sánchez-Eppler’s Touching Liberty (1993) traces how the same rhetoric of feeling that Tompkins credits with moving readers toward abolitionist sympathy did its work by making the enslaved body legible to a white reader chiefly as an object of that reader’s feeling, a suffering surface for a free woman’s tears rather than a person whose own interior life the novel troubled itself to render. The identification Dobson locates as the genre’s technical achievement is, on Sánchez-Eppler’s reading, also the mechanism of its worst failure: readers were moved to feel deeply for characters constructed to be felt for rather than known, and that construction persisted as a literary habit for decades after the cause it had served was won.

This does not return the argument to Douglas, because her complaint and Sánchez-Eppler’s are different kinds of charge wearing similar language. Douglas argues that sentimentalism was aesthetically and culturally soft, a symptom of decline in what American readers expected of their books. Sánchez-Eppler argues something narrower and harder to answer: that the genre’s own technique, exactly as sophisticated as Tompkins and Dobson describe it, achieved its effects by borrowing a body it had no standing to speak for. A rhetoric can be precisely as accomplished as its defenders claim and still be doing something worth objecting to. Craft and complicity sit on different axes, and a defence adequate to the first says nothing at all about the second.

What survives is narrower than either the traditional dismissal or the revisionist answer to it. Sentimental fiction should be read as a genre with real, locatable technical resources, aimed with some precision at readers who could be relied upon to respond to them in a predictable way. That fact settles nothing about whether the ends those resources served, or the people they were used on, deserve the response the genre was built to produce. Vindicating the prose and vindicating its politics are separate undertakings, and the century of argument over whether sentimental fiction is good has mostly proceeded as though winning one settled the other.

Little Eva’s death, then, is neither the evidence of incompetence Douglas’s heirs still cite nor the vindicated masterstroke Tompkins’s heirs sometimes make it out to be. It is a passage doing precisely what its genre’s techniques were built to do, to an audience whose response to it can no longer be checked, in the service of an argument about a child the novel needed far more than it needed her to be a person.