It [Robert Lenox Kennedy’s donation] is a very curious memorial of the taste, manners, and lighter literature of the country from the period of 1783 until about or near 1830. Newport in those days was […] a resort during the warm season of the most refined, wealthy, and cultivated society of the whole of this country; and this [James Hammond’s] Circulating Library was long the chief resort of that society for lighter literature […]

This circulating library contained the works of light reading at the time in most demand, including the numerous novels from the then fashionable Minerva Press in London, most of which have now passed away and are forgotten, although a few of them are by authors, like Mrs. Radcliffe, of great power and originality. [1]

In 1811, the Newport Mercury, a local newspaper in Newport, Rhode Island, reported that James Hammond, a dry-goods merchant, had purchased the stock of Wanton & Rathbone’s Providence Circulating Library. According to the article, Hammond’s ‘collection of Novels and Romances, is more extensive than can be found in any other Library in America’. [2] Operating his library for over half a century, Hammond grew his collection over time. In 1817, the Newport Mercury ran an advertisement claiming that the library offered over 3000 volumes; in 1858, a newspaper advertisement cited 10,000 volumes. [3] Not only nineteenth-century New England’s largest circulating library, Hammond’s library also became, by all accounts, an important institution in Newport and beyond. [4] In 1848, Henry Barnard claimed that ‘Hammond’s Circulating Library is one of the oldest libraries in the country’; in 1866, on the occasion of Hammond’s death, the Newport Mercury remarked that the library had served a ‘public benefit’. [5] When the library later closed its doors and its holdings went to auction, Robert Lenox Kennedy purchased approximately 1850 volumes and donated them to the New York Society Library, for which he served as trustee. [6]

Considering its size, its importance and its multi-decade operation, Hammond’s circulating library is a compelling site for exploring the fate of Minerva Press gothic novels on American shores decades after their publication. Gothic novels printed at William Lane’s Minerva Press account for only a small portion of the New York Society Library’s Hammond Collection: sixty-five novels published between 1791 and 1810. [7] However, the library’s 1868 trustees’ report, partially quoted in my epigraph, called attention to those novels as a noteworthy subset. The report associated the Minerva Press with Ann Radcliffe, conflating the press whose output included gothic novels with the gothic novelist herself, despite the fact that none of her novels were originally printed at that press (p. 5). It is true that the Minerva Press later reprinted two of Radcliffe’s novels, but, of those two, only The Romance of the Forest (1791) appeared in Hammond’s last-known extant catalogue (1858), and that novel was apparently an American edition. [8] The erroneous association between popular press and gothic novelist notwithstanding, the report nevertheless implied that Minerva Press gothic novels enjoyed a lively readership in nineteenth-century Newport. In fact, despite their dubious reputation, these novels counted among those that were ‘at the time’, the report claimed, ‘in most demand’ (p. 5). [9]

This essay investigates Hammond’s promotion of Minerva’s gothic novels—and Newport readers’ engagement with them—for almost half a century after the Minerva Press ceased to print under that name. In so doing, this essay contributes to scholarship that has shifted attention from the gothic’s initial publication and first readers to the gothic’s persistent circulation and later readership. Literary histories of the gothic novel that emphasise production over circulation chart the mode’s decline in the 1820s. [10] However, as Franz J. Potter and Edward Jacobs have argued in their work on early British gothic publishing and circulating libraries, a novel’s circulation and readership may extend well beyond its initial publication. [11] In fact, as Potter contends, ‘circulating libraries and their readers […] construct[ed] a different “canon” than is suggested by literary history’. [12]

Hammond, the proprietor of an important nineteenth-century American circulating library, contributed to the formation of just such an alternative canon. His contributions followed in the footsteps of Lane’s earlier ‘transcontinental and transatlantic network of printers, circulating libraries, and booksellers’. [13] A notable member of Lane’s network, Hocquet Caritat—turn-of-thenineteenth-century New York-based librarian, bookseller and publisher—held many Minerva Press titles, reprinted Regina Maria Roche’s Minerva title The Children of the Abbey (1796), introduced Charles Brockden Brown’s novels to Lane, and became the latter’s American agent in 1800. [14] More modest in his endeavours, Hammond nevertheless held a significant collection of Minerva Press gothic novels until at least 1858 but probably until the library went out of business approximately a decade later. Considering a circulating librarian’s motive to turn a profit, Hammond would not have retained these novels so long if he had not recognised their demand, nor would he have used his catalogues to promote them. [15] Hammond’s catalogues arranged and altered Minerva Press titles and authors and appended reviews. These promotional strategies, examined in section one, foregrounded the gothic content of these novels and attested to their quality. In so doing, the catalogues assisted subscribers in selecting books and prepared them for reading those selections.

Hammond’s patrons also played a role in keeping alive the novels’ ‘“experiential popularity”’. [16] Little is known about these readers, who resided in Newport or visited the resort town. If living in or travelling from the region, Hammond’s subscribers probably enjoyed a high literacy rate, read European and increasingly American literature and accessed novels from local circulating libraries. [17] Beyond those regional characteristics, these readers reportedly demanded Minerva Press gothic novels, apparently had access to them and likely borrowed and read them. Many of the volumes that they would have selected contained copious marginalia. Written by other subscribers or earlier readers, these marginalia, examined in section two, evaluated the novels and commented on their gothic and non-gothic elements alike. [18] In so doing, the marginalia sometimes supported, and at other times conflicted with, Hammond’s promotional strategies. If Hammond’s catalogues prepared patrons for quality gothic reading experiences, the novels’ marginalia reshaped patrons’ expectations and influenced their engagement with the novels.

For Hammond’s patrons, the catalogues and the marginalia constituted two points of entry into the Minerva Press gothic novel. Collectively but at times divergently, these paratexts ‘ensure[d]’ the Minerva Press gothic novel’s ‘“reception”’ in Newport and its ‘consumption in the form […] of a book’. [19] Neither paratext should be considered alone. Only by examining both the catalogues and the marginalia can we assess the degree to which the Minerva Press gothic novel terrified and delighted Newport readers decades after its London heyday.

Catalogues: Promoting the Minerva Press Gothic Novel

‘Persons Sending for Books from this Catalogue, will Please to send by the Numbers’, requested Hammond’s 1858 catalogue. Hammond published this catalogue and others not only to assist patrons in locating and selecting books but also, and more importantly, to promote his stock. As one of his primary promotional tools, the catalogues affected how patrons encountered the library’s collection of Minerva Press gothic novels. According to Edward Jacobs, circulating-library catalogues ordered literary knowledge in highly predictable ways. In so doing, they ‘called upon book selectors to perceive, experience, and judge books in ways that interacted […] with the practices of reading and writing’. [20] Contending that Hammond’s catalogues performed a similar function, this section considers four catalogues with particular attention to the 1858 catalogue, which not only contained all of Minerva’s gothic novels under examination but also reflected Hammond’s cataloguing strategies towards the end of the library’s operation. [21] The catalogues organised, altered and supplemented many of the collection’s Minerva gothic titles. As a result, the catalogues singled out the novels, highlighted their gothic mode and provided evidence of their quality.

All four catalogues arranged titles in alphabetical order in the main or a generic listing. According to Garside and Jacobs, that organisational strategy encouraged generic reading practices. Period novels, notoriously gothic novels, recycled the same title lead words, such as ‘castle’ and ‘mystery’. As a result, similar titles inevitably appeared alongside each other in a catalogue’s alphabetical listing. Subscribers who wanted novels akin to those they had already read would have likely turned to titles with the same title lead word even if that presumed generic similarity was misleading. [22] Enhancing the impression that novels with similar title lead words were virtually interchangeable, Hammond’s library occupied the same space as his dry-goods store, as lists of available dry goods in the 1820s and 1844–45 catalogues reminded subscribers. The spatial proximity of books and dry goods, two commodities operating on different business models, invited customers to believe that one castle novel could easily substitute for another just as one piece of linen might serve as well as another.

Although it is generally the case that Hammond’s cataloguing strategies and business architecture promoted just such a perception, the 1858 catalogue arranged some novels more complexly. For example, when hypothetical subscribers opened the catalogue, they would have discovered eighteen titles beginning with ‘mystery’ or some variation. While it may have given the impression that mystery novels were essentially interchangeable, this arrangement alone in no way guaranteed that all eighteen novels belonged to the same genre or that, moreover, subscribers would have recognised among them five original Minerva Press gothic titles. Curiously, Hammond placed these five titles out of strict alphabetical order (to which he rarely conformed) in an almost continuous, nearly uninterrupted numeric sequence in the 1300s within the larger mystery section, whose numbers ranged from the 1300s to the 6000s. In all likelihood, the Minerva sequence corresponded to shelf order and reflected the novels’ acquisition dates. In any case, the catalogue did not organise titles numerically, and other, non-Minerva titles that should have belonged to this sequence by virtue of number appeared elsewhere in the mystery section. Thus, even if it resulted from an alphabetical or a numerical accident, this cluster nevertheless called attention to these Minerva Press gothic novels as a particular subgenre of mystery novel while effectively preventing them from being dispersed across the section. Augmenting this visual effect, the catalogue placed just before the Minerva sequence Radcliffe’s The Mysteries of Udolpho (1794), the imminently recognisable gothic novel. That placement showcased these Minerva Press gothic novels as a worthy subdivision of Hammond’s collection of otherwise potentially fungible mystery novels.

In addition to arranging Minerva Press titles, Hammond occasionally altered them. In fact, circulating-library proprietors often enhanced titles to inform and entice readers. [23] Hammond’s alterations typically foregrounded novels’ genres and settings. For example, he changed some titles’ genres from ‘novel’ to ‘romance’ and even added nationalities to those generic designations. In the 1858 catalogue, the genres of The Castle of Ollada. A Romance (1795), Lusignan, or the Abbaye of La Trappe. A Novel (1801) and Edmund of the Forest. An Historical Novel (1797) (as rendered on the novels’ title pages [24]) became, respectively, ‘a Spanish romance’, ‘a French romance’ and ‘a romance’. By replacing ‘novel’ with ‘romance’, Hammond signalled to patrons the titles’ gothic genre. [25] Considering his patrons’ alleged taste for the Minerva Press gothic brand, Hammond had a financial motive in making the novels’ genre explicit. Moreover, his addition of national markers invoked far-flung, even exotic settings that might have appealed to thrill-seeking Newport readers. Thus, collectively, these altered titles served not only ‘to designate’ and ‘to indicate subject matter’, to borrow Genette’s words, but also ‘to tempt the public’. [26]

While Hammond most often organised and altered Minerva Press titles, he occasionally changed their authorship, notably to novels published anonymously. In fact, almost half of the Minerva Press gothic novels in Hammond’s collection were published anonymously—a statistic roughly in line with Minerva Press authorship in 1800. [27] Pervasive during the Romantic period, anonymous-novel publishing served ‘to impose greater similarity, regularity, and predictability on the nature and habit of novel-reading, in other words to turn novels into uniform and mutually substitutable commodities, and the renting and reading of them into a regular habit’. [28] If novels could be interchanged by virtue of anonymous authorship or similar title, they risked getting lost in the shuffle of such a large circulating library. To avoid that consequence, Hammond could have identified the authors, when known, of anonymous novels and elevated them to the select company of authors whose titles he listed in separate author sections in the three 1840s and 1850s catalogues. Undoubtedly, the author sections drew subscribers’ attention to certain novelists, such as Walter Scott, thirty-nine of whose titles appeared under his name in the 1858 catalogue’s author section. However, that section listed only one author, Eliza Parsons, with at least one Minerva gothic novel in the subset under examination, and that novel, The Mysterious Warning (1796), was not published anonymously.

Instead, Hammond chose other strategies, ones admittedly tenuous and thus less effective but ones that nonetheless used authorship for their ‘classificatory function’: a function that, according to Michel Foucault, ‘permits one to group together a certain number of texts, define them, differentiate them from and contrast them to others’. [29] For example, the 1858 catalogue included Minerva Press gothic novels originally attributed to Anna Maria Mackenzie and ‘Gabrielli’, the pseudonym of Elizabeth Meeke. In addition, the catalogue listed Danish Massacre (1791) and The Sicilian (1798), novels attributed at publication by way of the authors’ previous novels but attributed by Hammond, respectively, to Mrs Mackenzie and Gabrielli. [30] Although several Mackenzie and Gabrielli novels were dispersed throughout the main title listing, fans would have been more apt to select the novelists’ other titles if attributed by name. Hammond used a different technique with Melbourne (1798). While the title page attributed the novel to ‘the author of Deloraine [1789]’, he credited ‘the author of Reginald, &c.’ Hammond likely took his cue from the title page of Reginald, ‘by the author of Melbourne, &c.’ By substituting Reginald for Deloraine—an ‘Italian romance’ (as his catalogue labelled the former) for ‘a domestic tale’ (as his catalogue and the title page labelled the latter); a novel attributed to Melbourne’s author for an anonymous novel—Hammond effectively cross-referenced two Minerva gothic novels. [31] Though subtle, these changes grouped some anonymous Minerva Press gothic novels and, in some cases, linked them to other novels attributed by name. These groupings effectively drew subscribers’ attention to particular authors.

Generally, Hammond’s arrangements and alterations informed subscribers about book content while enticing them to select certain novels. Supplementary information in some catalogue entries, notably in the 1820s catalogue, not only informed and enticed readers but also evaluated the novels. A few entries offered unattributed quotations, sometimes derived from the novels’ title pages. For instance, the quotation appended to the catalogue entry for The Mystic Castle (1796) abbreviated and spliced the two Hamlet epigraphs on the novel’s title page: ‘“Foul deeds will rise; murder will speak”’ (p. 47). [32] Though compressed, this literary quotation gave subscribers with a penchant for the gothic an immediate impression of the novel’s lurid themes. To enhance just such an effect, the catalogue passed over the epigraph on the title page of The Mysterious Warning to capitalise on the novel’s gothic theme of mystery. Instead of ‘“Thus conscience | Can make cowards of us all”’, [33] the catalogue entry read: ‘“a seared conscience fears mystery”’ (p. 47). Other entries in the 1820s catalogue provided information, sometimes quoted and attributed, that likely came from the novels’ contemporary reviews. For example, the catalogue quoted from a review, attributed to the Critical Review, of The Forest of Hohenelbe (1803): ‘here are caverns inhabited by outlaws, massy chains, and impenetrable prisons: all probable’ (p. 38). If the title alone did not sufficiently reveal the novel’s content, the excerpted review detailed the gothic elements that awaited the book selector. The excerpted review of Melbourne, quoted and attributed to the Critical Review, forewent the gothic for a simple declaration of the novel’s quality: ‘“the characters are original and interesting, the plan ingenious”’ (p. 45). For subscribers unfamiliar with particular Minerva Press gothic novels in Hammond’s collection, reviews such as this one assured book selectors of the novels’ quality.

In appending these reviews, Hammond made an unusual marketing decision. Certainly, his decision was not without precedent. For example, Caritat included excerpts from London reviews in his so-called ‘explanatory catalogs’. [34] However, by the 1820s, at least one to three decades after the publication of Hammond’s Minerva gothic novels, these recycled British reviews surely appeared outdated to Newport readers. Rather than updating them to reflect then-current American tastes, Hammond mostly abandoned this marketing strategy in later catalogues. Even so, Hammond’s subscribers still had access to reviews that the Minerva Press appended to advertisements in the volumes’ final pages. [35] For example, at the end of Volume Three of Anecdotes of the Altamont Family (1800), the advertisement for Reginald, or the House of Mirandola (1799), ‘Just published’, quoted a flattering excerpt from the London Review, dated January 1800. [36] Although Hammond obviously did not insert these advertisements, they nevertheless informed readers about the novels’ content and their perceived quality. In so doing, the advertisements supplemented Hammond’s promotional strategies—but probably ineffectively and inauthentically: ineffectively because, as remnants of a then-defunct London press, they solicited mid-century Newport patrons to read, even to purchase, novels oddly described as ‘just published’; inauthentically because the positive reviews came from earlier professional British readers. In contrast, the volumes’ marginalia, to which this essay now turns, provided advice from ordinary readers about how to read these novels.

Marginalia: Modelling the Reading Experience

‘Books lost, written in, or otherwise damaged, must be paid for’, warned Hammond’s subscription terms, pasted in Orwell Manor (1795) and other novels. [37] Not alone in condemning marginalia, Hammond’s library conspired with other nineteenth-century institutions, such as schools and public libraries, to curb the habit. Despite the warning, some readers left, or had already left, copious marginalia in Hammond’s Minerva Press gothic collection. Like their predecessors across the centuries, these writers of marginalia sought to improve, correct, supplement, cross-reference, assess, correct and otherwise comment on texts. With only a few possible exceptions, these marginalia, ‘the actual responses of actual readers’ at some time and in some place, cannot be definitively attributed to Hammond’s patrons, since his collection probably belonged to at least two previous libraries. [38] Regardless of their origins, these marginalia nevertheless reflected and shaped an ‘interpretative community’ of Minerva gothic-novel readers, including Hammond’s Newport subscribers across a half century. [39] These marginalia sometimes underscored and at other times undermined Hammond’s promotional strategies. In so doing, the marginalia served as models for how to read Minerva Press gothic novels: models that Hammond’s patrons could consider and either adopt or reject. [40]

Many readers commented on characters and plots. More than any other type of comment, these comments illustrate the differences between Hammond’s catalogues and readers’ marginalia. Preparing subscribers for gothic reading experiences, the 1858 catalogue placed The Mystic Castle and The Castle of Caithness in gothic company with, respectively, other ‘mystery’ and ‘castle’ listings, and the 1820s catalogue quoted parts of the eerie epigraph on The Mystic Castle’s title page. However, the readers of these novels did not limit their attention to gothic elements. For example, one reader of The Mystic Castle labelled Jane, De Mowbray’s aspiring daughter, ‘a hateful bitch’ (II, 36). One reader of The Castle of Caithness provided a playful running commentary on a passionate scene between the imprisoned Isabel and Edward. Next to passages in which the characters swoon and embrace, the reader wrote: ‘how I do love her’; ‘and then what fun’; ‘Oh what loving times’; ‘dear good little soul’; ‘and kisses him’. [41] While those readers shifted attention from the gothic, one reader of Reginald engaged with the gothic only to deflate it. At one point, Sigismond, the rightful heir of Mirandola, sees a tombstone engraved with his mother Hypolita’s name. Underneath the inscription, the reader wrote: ‘She is still living’ (III, 203). This plot spoiler joined the other readers above in modelling reading experiences that read away from the gothic or against it altogether. As a result, they compromised Hammond’s promotional goals and invited later readers to follow suit.

Like commenters on characters and plots, readers who supplemented the volumes with visual and textual materials engaged with non-gothic and gothic features alike. Their supplements ranged from the whimsical and the parodic to the studious. For example, in the first few blank pages of Volumes Three and Four, a reader of Correlia, or the Mystic Tomb (1802) provided makeshift frontispieces. Drawing as many as four illustrations, the reader labelled two of them ‘beautiful Correlia’. [42] At the end of Lusignan, or the Abbaye of La Trappe, a reader added advertisements below and opposite the existing advertisement for First Love. A Novel. In Three Volumes (1801). Cleverly mimicking print, this reader created advertisements for ‘Second Love in Four Volumes’ and for ‘Nothing in Eighty Volumes’ (IV, n.p.). By parodying its predictable titles, its multi-volume formats and even its perceived vacuousness, this reader undermined the literary pretensions of Minerva Press novels and arguably influenced subsequent readers to adopt a similar opinion. In contrast, other readers appreciated the novels’ literary aspirations. For instance, in the first volume of Edmund of the Forest, Edmund requests that his host, Sir James, ‘listen to my tale’. Apparently detecting an allusion to Hamlet (I. 5. 15), one reader underlined ‘I shall unfold’ (the phrase that followed) and noted at the foot of the page: ‘I shall a tale unfold, &c’ (p. 222). Unlike Hammond, who lifted lines from title pages for promotional purposes, this reader recognised the novel’s literariness, demonstrated literary competence for the gothic in particular and facilitated future readers’ appreciation for the literary mode.

Readers who edited the novels also engaged with literary language, not necessarily to appreciate it, but often to improve, parody and even criticise it. Take, for example, a series of edits to Edmund of the Forest. In Volume Two, a reader struck the phrase ‘thought he’ towards the beginning of a passage about Edmund’s internal reflections and placed the phrase at the end of the paragraph (p. 63). Striking the word ‘ah’ in that passage, a reader, probably the same one, wrote ‘Too many ah’s’ next to an advertisement at the end of Volume Two and later, in the first few pages of the next volume, added ‘B’ to ‘Ah’ as well as ‘Ah!’ and ‘O, Ah!’ to numerous passages (pp. 10–15). Back in Volume Two, the same or another reader wrote below an epigraph ‘Too much of’ next to ‘Shakespeare’ (p. 110). In the next volume, four of the first seven chapters begin with Shakespeare epigraphs. Arriving at Chapter 8 and discovering an epigraph from John Home’s tragedy Douglas (1756), the weary reader wrote ‘Good’ above the epigraph’s attribution and changed the epigraph’s ‘ah!’ to ‘oh!’ (p. 113). In making these edits, the reader(s) demonstrated and modelled a different level of literary competence from that of the reader who detected the Hamlet allusion. Showing sufficient literary sophistication to recognise an author’s repetitive patterns and tiresome homage to Shakespeare, the reader(s) questioned the quality of the Minerva Press gothic novel while suggesting that at least some readers no longer expected deference to the literary canon. [43]

In commenting on, supplementing and editing the novels, earlier readers surely anticipated a future readership for their marginalia. However, it is mainly when they evaluated the novels’ quality that we witness their creation of a larger ‘interpretative community’ and their implicit invitation to future readers to join. For instance, when the ghost of the murdered Glencairn appears to his son Edward in The Castle of Caithness, one B declared: ‘I like not the gosts [sic] in romances.’ In response, another reader wrote ‘not me either’. Joining the running commentary, one L took a shot at B’s orthographic deficiencies: ‘I wish B knew how to spell Ghost’ (II, 73). Two of these comments showed a lack of appreciation for the novel’s gothic machinery; two comments engaged as much with other readers as with the novel itself. Broadening their scope to the novel as a whole, multiple readers left overwhelmingly favourable comments at the end of the novel: ‘Highly animated material and engaging’; ‘Excellent’; ‘Beautiful’; ‘Superlative in the highest degree’ (II, 256). This sequence and others like it in Hammond’s collection prove that some readers noticed previous readers’ comments. Moreover, unlike professional reviews, these marginalia reflected the musings of ordinary readers who not only evaluated the novels but also left space for others to contribute. This makeshift ‘interpretative community’ of unknown readers potentially included future readers who, even if they left no trace, still encountered, and were arguably influenced by, these marginalia.

Obviously intended and promoted for reading, these novels often met unanticipated fates in the hands of some readers. For instance, Romance of the Castle (1800) served, in part, as notepaper. On a blank page at the front of Volume One, a reader listed three Walter Scott novels: ‘Rob Roy Guy Mannering Waverly [sic]’. [44] Whether dissatisfied with Romance of the Castle or simply planning future reading, this reader ironically used the paper in an anonymous popular novel printed at a down-market press to list a celebrity author’s respectable novels. [45] On a blank page at the back of Volume Two, a reader (likely one Stephen, whose name appears on the same page) recorded a palindrome by John Taylor: ‘Lewd did I live and evil [I did] dwel’. Perhaps the evil machinations of the novel’s Longueville inspired the notation, or perhaps the notation had no connection to the novel’s plot at all. In any case, Stephen, like the list-maker, valued the novel as much for its material as for its content and invited subsequent readers to entertain similar uses that Hammond would not have intended, much less promoted.

All in all, these sample marginalia extended and countered Hammond’s promotional strategies in both purpose and content. Generally speaking, Hammond’s catalogues informed subscribers and enticed them to select novels by highlighting the novels’ gothic characteristics while promising quality reading experiences. In contrast, the marginalia read with, against and beyond the gothic; they accepted and questioned the novels’ literary quality; they even put the novels to alternative uses. In leaving traces of what they read, how they read it and how they used books, these unknown readers shaped the reading experiences of future readers, including Hammond’s subscribers. Clearly, this community was far from univocal in its appraisal of the Minerva Press gothic novel. Two comments in particular epitomised readers’ divergent views. One reader of The Mystic Castle teased future readers by writing upside down and backwards: ‘Novels are bad things Don[’]t you think so’ (I, 131). A reader of Who’s the Murderer? Or the Mystery of the Forest (1802) could not have disagreed more: ‘He that says there is no Good from a Novel is a Liar’. [46]

Conclusion

James Hammond’s Circulating Library and, in particular, its collection of Minerva Press gothic novels illuminate one aspect of nineteenth-century Newport experience. The term ‘experience’ derives from the late work of Foucault. In his 1982–83 lectures at the Collège de France, he defines ‘focal points of experience’ as the intersection of ‘a possible knowledge (savoir), normative frameworks of behavior for individuals, and potential modes of existence for possible subjects’. [47] The first two ‘dimensions of […] experience’—namely, knowledge and power—found expression in Hammond’s promotion of Minerva Press gothic novels. [48] Using catalogues as one of his primary promotional tools, Hammond organised literary knowledge by title and other recognisable terms. Within those arrangements, he occasionally enhanced individual catalogue entries with reviews and alternative titles. The catalogues in turn shaped Hammond’s patrons by informing their literary knowledge, influencing their book selection and moulding their expectations about their upcoming reading experiences.

The marginalia constitute the third ‘dimension’ of this particular Newport ‘experience’—in brief, subjectivity. Subject formation is, Foucault implies, two-fold: both a process that potential subjects undergo and the models that they consider while undergoing that process. Unfortunately, we cannot definitively determine the models of subject formation that influenced this study’s marginalia writers. However, from their marginalia, we see the effects of their formation as readers of Minerva Press gothic novels, a specific subjectivity among others available to Newport readers of circulating-library books. According to Foucault, ‘the individual’ who undergoes a process of subject formation ‘delimits that part of himself that will form the object of his moral practice, defines his position relative to the precept he will follow, and decides on a certain mode of being that will serve as his moral goal’. [49] Likewise, these marginalia writers staked out ‘position[s]’ towards literary language: whether to read for enjoyment or for literary competence—or whether to read at all. Likewise, they adopted ‘mode[s] of being’: attitudes—playful, derisive and even parodic—towards characters, actions and words; positions of editor and evaluator; and relationships—critical or agreeable and almost certainly imaginative—with unknown readers across time.

These writers’ marginalia in turn served as models for future readers, including Hammond’s patrons, in their own process of subject formation. These models at times supplemented, and at other times diverged from, Hammond’s promotional strategies, which initially shaped patrons’ expectations of the novels’ content. According to Foucault, ‘models’ of subject formation include ones ‘proposed for setting up and developing relationships with the self’ and ‘for the transformations that one seeks to accomplish with oneself as object’. [50] At a basic level, the marginalia provided options for what and how to read—options that subscribers could take, ignore or modify. In making those concrete decisions, however small and seemingly inconsequential, readers effectively enhanced and even altered their ongoing formation as Minerva gothic readers.

To understand this nineteenth-century Newport ‘experience’, we must examine both Hammond’s catalogues and readers’ marginalia. Without addressing both paratexts, we would fail to see what induced these resort-town readers to read Minerva Press gothic novels and how a library proprietor and earlier readers shaped future gothic reading experiences. Such an analysis necessarily requires attention to a specific historical site, and that focused attention obviously comes with limitations. After all, James Hammond’s Circulating Library—its catalogues, its collection and its patrons—cannot tell us everything about nineteenth-century American readers of Minerva Press gothic novels. That limitation notwithstanding, this local site nevertheless increases our understanding of the transatlantic, multi-decade history of Minerva Press gothic novels while suggesting similar sites for future work.

Notes

For opening their collections and providing research space, I thank the New York Public Library, the Grolier Club, the Morgan Library and Museum and especially the New York Society Library. Special thanks go to Barbara Bieck and Laura O’Keefe, New York Society Library staff members, for their expert assistance and patient support. For their invaluable suggestions on this essay, I thank Christina Morin, Elizabeth Neiman and the anonymous peer reviewer.

1. New York Society Library, Annual Report of the Trustees, March 30, 1868 (New York: New York Printing Co., 1868), pp. 4–5. Further references are cited parenthetically in the text. The following guided me to the report: Austin Baxter Keep, History of the New York Society Library (Boston: Gregg Press, 1972), pp. 471–72.

2. Newport Mercury, 2 Feb 1811, in America’s Historical Newspapers <https:// www.readex.com/content/americas-historical-newspapers> [accessed 9 Jan 2018]. Further references are cited in the text; further references to newspapers come from this database. The following guided me to newspaper articles in this paragraph: H. Glenn Brown and Maude O. Brown, A Directory of Printing, Publishing, Bookselling, and Allied Trades in Rhode Island to 1865 (New York: New York Public Library, 1958), pp. 80–81. For helpful histories of Hammond’s library, see David Kaser, A Book for a Sixpence: The Circulating Library in America (Pittsburg: Beta Phi Mu, 1980), pp. 97–98 and 142; Jesse H. Shera, Foundations of the Public Library: The Origins of the Public Library Movement in New England 1629–1855 (Chicago: University of Chicago Press, 1949), p. 146. Kaser and Shera guided me to many sources in this paragraph.

3. Newport Mercury, 10 May 1817; Newport Mercury, 10 July 1858.

4. Kaser, Book for a Sixpence, p. 97.

5. Report and Documents Relating to the Public Schools of Rhode Island for 1848 (Providence, RI: [n. pub.], 1849), p. 426, in Sabin Americana, 1500–1926 <http:// galenet.galegroup.com/servlet/Sabin?af=RN&ae=CY110329583&srchtp= a&ste=14> [accessed 3 Jan 2018]; Newport Mercury, 7 July 1866.

6. George Champlin Mason, Reminiscences of Newport (Newport, RI: Hammett, 1884), pp. 275–76; New York Society Library, p. 4.

7. This subset was determined by consulting the New York Society Library’s online catalogue (also the source for subsequent references to the Hammond Collection) and two standard bibliographies of British gothic novels: Maurice Lévy, Le Roman ‘gothique’ anglais 1764–1824 (Paris: Michel, 1995); Frederick S. Frank, The First Gothics: A Critical Guide to the English Gothic Novel (New York and London: Garland, 1987). A few discrepancies exist across the lists and indexes within each bibliography; if a novel appears in at least one of those places, it was counted. Also consulted was Peter Garside, James Raven and Rainer Schöwerling, The English Novel 1770–1829: A Bibliographical Survey of Prose Fiction Published in the British Isles, 2 vols (Oxford: OUP, 2000). Garside, ‘The English Novel in the Romantic Era: Consolidation and Dispersal’, in English Novel, II, 15–103 (pp. 55–56) cites Lévy and Frank as two standard gothic bibliographies.

8. For Minerva reprints of Radcliffe, see Montague Summers, A Gothic Bibliography (London: Fortune Press, [1941]), pp. 483 and 503. See also Catalogue of James Hammond’s Circulating Library (Newport, RI: Power Press, 1858), in Reading: Harvard Views of Readers, Readership, and Reading History <http://ocp.hul. harvard.edu/reading/> [accessed 4 Jan 2018] and the New York Society Library’s online catalogue. Further references to the 1858 catalogue are cited in the text.

9. For discussions of negative views of the Minerva Press, see e.g. John Tinnon Taylor, Early Opposition to the English Novel: The Popular Reaction from 1760 to 1830 (New York: King’s Crown Press, 1943), p. 28; E. J. Clery, The Rise of Supernatural Fiction, 1762–1800 (Cambridge: CUP, 1995), pp. 135–40; Elizabeth Neiman, ‘The Female Authors of the Minerva Press and “Copper Currency”: Revaluing the Reproduction of “Immaculate-Born Minervas”’, in Global Economies, Cultural Currencies of the Eighteenth Century, ed. by Michael Rotenberg-Schwartz with Tara Czechowski (New York: AMS Press, 2012), pp. 275–94 (pp. 275–78).

10. Franz J. Potter, The History of Gothic Publishing, 1800–1835: Exhuming the Trade (Basingstoke and New York: Palgrave Macmillan, 2005), pp. 1–13; Robert Miles, ‘The 1790s: The Effulgence of Gothic’, in The Cambridge Companion to Gothic Fiction, ed. by Jerrold E. Hogle (Cambridge: CUP, 2002), pp. 41–62 (pp. 42–43); Garside, ‘English Novel in the Romantic Era’, p. 56.

11. Potter, History of Gothic Publishing, pp. 9–10, 15 and 21–36; Edward Jacobs, ‘Eighteenth-Century British Circulating Libraries and Cultural Book History’, Book History, 6 (2003), 1–22 (pp. 1–3 and 14–17) <https://doi.org/10.1353/bh.2004.0010>.

12. Potter, History of Gothic Publishing, p. 36.

13. Christina Morin, ‘Irish Gothic Goes Abroad: Cultural Migration, Materiality, and the Minerva Press’, in Traveling Irishness in the Long Nineteenth Century, ed. by Marguérite Corporaal and Christina Morin (Basingstoke and New York: Palgrave Macmillan, 2017), pp. 185–203 (p. 194).

14. See ibid., pp. 192–93; Eve Tavor Bannet, ‘Charles Brockden Brown and England: Of Genres, the Minerva Press, and the Early Republican Reprint Trade’, in Transatlantic Literary Exchanges 1790–1870: Gender, Race, and Nation, ed. by Kevin Hutchings and Julia M. Wright (Farnham: Ashgate, 2011), pp. 133–52 (esp. pp. 134–35); George Gates Raddin, Jr, Hocquet Caritat and the Early New York Literary Scene (Dover, NJ: Dover Advance Press, 1953), pp. 29, 58, 82, 89 and 122; Kaser, Book for a Sixpence, pp. 56–60.

15. For discussions of the relationship between consumer demand and circulating librarians’ profit motives and catalogue listings, see Potter, History of Gothic Publishing, p. 22; Shera, Foundations of the Public Library, p. 155; Jacobs, ‘Eighteenth-Century Circulating Libraries’, pp. 1–3 and 14–17; Edward Jacobs, ‘Anonymous Signatures: Circulating Libraries, Conventionality, and the Production of Gothic Romances’, ELH, 62 (1995), 603–29 (p. 605); Edward Jacobs, Accidental Migrations: An Archaeology of Gothic Discourse (Lewisburg and London: Bucknell University Press and Associated University Presses, 2000), p. 159.

16. The phrase derives from Jacobs, ‘Eighteenth-Century Circulating Libraries’, p. 15.

17. See e.g. Kaser, Book for a Sixpence, pp. 50, 62–68, 86 and 98–103; Shera, Foundations of the Public Library, pp. 127 and 133–55; Richard D. Brown, Knowledge Is Power: The Diffusion of Information in Early America, 1700–1865 (New York and Oxford: OUP, 1989), p. 12; Cathy N. Davidson, Revolution and the Word: The Rise of the Novel in America, new edn (New York: OUP, 2004), pp. 3–4, 7, 68, 87–90, 119 and 123; William St Clair, The Reading Nation in the Romantic Period (Cambridge: CUP, 2004), pp. 382–89; James L. Machor, ‘Introduction: Readers/Texts/Contexts’, in Readers in History: Nineteenth-Century American Literature and the Contexts of Response, ed. by James L. Machor (Baltimore and London: Johns Hopkins University Press, 1993), pp. vii– xxix (p. xii); Haynes McMullen, American Libraries before 1876 (Westport, CT and London: Greenwood Press, 2000), pp. 137–39.

18. Hammond’s Minerva gothic novels likely came from one or both of his immediate predecessors in Rhode Island: William R. Wilder, who set up a library in 1798 and sold it in 1806 to Wanton & Rathbone, who then sold their collection to Hammond in 1811. See Kaser, Book for a Sixpence, p. 97. At least one novel in the subset bears subscription terms from earlier libraries. See ‘F. H. P.’, The Castle of Caithness. A Romance of the Thirteenth Century, 2 vols (London: Minerva Press, 1802). Both volumes of this novel have been rebound. Attached to the inside front cover of vol. 1 are subscription terms for Hammond’s library and Henry Cushing’s Circulating Library in Providence; attached to the inside front cover of vol. 2 are subscription terms for those two libraries and for Wanton & Rathbone’s Circulating Library. This novel thus gives evidence of a third circulating library to which at least two of Hammond’s volumes belonged. All Minerva Press gothic novels cited in this article are held in the New York Society Library.

19. Gerard Genette, Paratexts: Thresholds of Interpretation, trans. by Jane E. Lewin (Cambridge: CUP, 1997), p. 1. Catalogues and marginalia meet Genette’s definition of ‘paratext’: material that lives outside the physical volume (epitext) and material bound in it (peritext); see pp. 4–5.

20. Edward Jacobs, ‘Buying into Classes: The Practice of Book Selection in EighteenthCentury Britain’, Eighteenth-Century Studies, 33 (1999), 43–64 (pp. 45 and 50–60) <https://doi.org/10.1353/ecs.1999.0052>; Edward Jacobs, ‘Ann Radcliffe and Romantic Print Culture’, in Ann Radcliffe, Romanticism and the Gothic, ed. by Dale Townshend and Angela Wright (Cambridge: CUP, 2014), pp. 49–66 (pp. 50 and 53–54); Jacobs,

Accidental Migrations, pp. 172–92. Jacobs has generally influenced my understanding of catalogues here and below.

21. The other catalogues consulted include: Catalogue of James Hammond’s Circulating Library (Newport, RI: [n. pub.], [182–?]); Catalogue of James Hammond’s Circulating Library (Newport, RI: [n. pub.], [1844–45]); Catalogue of James Hammond’s Circulating Library (Newport, RI: Power Press, 1853), in Google Books <http://books.google.com> [accessed 13 Jan 2019]. Further references to these catalogues are cited in the text usually without page numbers, as titles are listed alphabetically. I preserve Hammond’s idiosyncratic spellings and versions of titles and authors but, for brevity’s sake, typically shorten titles.

22. Garside, ‘English Novel in the Romantic Era’, p. 49 as well as Jacobs, ‘Anonymous Signatures’, pp. 616–17; Jacobs, ‘Buying into Classes’, p. 54; Jacobs, Accidental Migrations, pp. 180–81, 184–85 and 190–91; Jacobs, ‘Radcliffe and Romantic Print Culture’, pp. 50 and 54.

23. For market-driven title selections and title changes in circulating-library catalogues, see Garside, ‘English Novel in the Romantic Era’, pp. 20, 50 and 52. For the use of titles to entice readers, see Potter, History of Gothic Publishing, p. 17.

24. Francis Lathom, The Castle of Ollada. A Romance, 2 vols (London: Minerva Press, 1795); Anon., Lusignan, or the Abbaye of La Trappe. A Novel, 4 vols (London: Minerva Press, 1801); Agnes Musgrave, Edmund of the Forest. An Historical Novel, 4 vols (London: Minerva Press, 1797). Further references to these novels are cited parenthetically in the text.

25. For a discussion of the romance genre in the context of turn-of-the-nineteenthcentury British fiction produced by circulating-library publishers, see Jacobs, ‘Anonymous Signatures’, pp. 617–24.

26. Genette, Paratexts, p. 76.

27. Dorothy Blakey, The Minerva Press, 1790–1820 (London: Bibliographical Society, 1939), p. 48.

28. William St Clair, ‘Publishing, Authorship, and Reading’, in The Cambridge Companion to Fiction in the Romantic Period, ed. by Richard Maxwell and Katie Trumpener (Cambridge: CUP, 2008), pp. 23–46 (p. 40).

29. Michel Foucault, ‘What Is an Author?’, in Aesthetics, Method, and Epistemology, ed. by James D. Faubion and trans. by Robert Hurley and others, Essential Works of Foucault, 1954–1984, II (New York: New Press, 1998), pp. 205–22 (p. 210).

30. Anna Maria MacKenzie, Danish Massacre, an Historic Fact, 2 vols (London: Minerva Press, 1791); Elizabeth Meeke, The Sicilian. A Novel, 4 vols (London: Minerva Press, 1798).

31. Mrs Martin, Melbourne. A Novel, 3 vols (London: Minerva Press, 1798); Mrs Martin, Reginald, or the House of Mirandola. A Romance, 3 vols (London: Minerva Press, 1799). Further references to Reginald are cited parenthetically in the text. For Deloraine, see Garside, Raven and Schöweling, English Novel, I, 752.

32. Mr Singer, The Mystic Castle; or, Orphan Heir. A Romance, 2 vols (London: Minerva Press, 1796). Further references are cited parenthetically in the text.

33. Eliza Parsons, The Mysterious Warning, a German Tale, 4 vols (London: Minerva Press, 1796). The epigraph was taken from the title page of the second volume, as the first volume has been rebound.

34. Explanatory Catalogue of H. Caritat’s Circulating Library; No. 1. City Hotel, Broad-Way, New York ([New York]: Waite, [1803?]).

35. For discussions of the Minerva Press’s advertising techniques, see Blakey, Minerva Press, pp. 101–04; Clery, Rise of Supernatural Fiction, p. 136; Neiman, ‘Female Authors’, pp. 276–77.

36. Elizabeth Meeke, Anecdotes of the Altamont Family. A Novel, 4 vols (London: Minerva Press, 1800).

37. Mary Elizabeth Parker, Orwell Manor. A Novel, 3 vols (London: Minerva Press, 1795).

38. The quotation and the general information about marginalia come from H. J. Jackson, Marginalia: Readers Writing in Books (New Haven and London: Yale University Press, 2001), pp. 72–75 and 253; and H. J. Jackson, Romantic Readers: The Evidence of Marginalia (New Haven and London: Yale University Press, 2013), pp. 120, 251–56 and 299–301.

39. The term derives from Davidson, Revolution and the Word, p. 62.

40. Many of the marginalia analysed below are part of longer, sometimes partially illegible comments. For brevity, I quote only decipherable phrases relevant to my discussion.

41. ‘F. H. P.’, Castle of Caithness, II, 166–69. Further references are cited parenthetically in the text.

42. Sarah Sheriffe, Correlia, or the Mystic Tomb. A Romance, 4 vols (London: Minerva Press, 1802).

43. For the argument that gothic novelists selected epigraphs from the literary canon to legitimise the popular genre, see e.g. Jacobs, ‘Radcliffe’, pp. 57–61; Deidre Shauna Lynch, Loving Literature: A Cultural History (Chicago and London: University of Chicago Press, 2015), pp. 225–26 and 229–30.

44. Jane Elson, Romance of the Castle, 2 vols (London: Minerva Press, 1800).

45. Kaser, Book for a Sixpence, p. 64, and St Clair, Reading Nation, p. 388, address the popularity of Scott among nineteenth-century American readers.

46. Eleanor Sleath, Who’s the Murderer? Or the Mystery of the Forest. A Novel, 4 vols (London: Minerva Press, 1802), IV, 398.

47. Michel Foucault, The Government of Self and Others, Lectures at the Collège de France, 1982–83, ed. by Arnold I. Davidson and trans. by Graham Burchell (New York: Picador, 2011), p. 3.

48. Ibid., p. 4.

49. Michel Foucault, The Use of Pleasure, trans. by Robert Hurley, The History of Sexuality, 2 (New York: Vintage Books, 1990), p. 28. See pp. 3–32 for Foucault’s larger discussion of this point in the context of ‘experience’.

50. Ibid., p. 29.