Free from unwarranted superstition though I am, these things produced in me an odd sensation, which was intensified by a pair of widely separated newspaper cuttings relating to deaths in the shunned house – one from the Providence Gazette and Country-Journal of April 12, 1815, and the other from the Daily Transcript and Chronicle of October 27, 1845 – each of which detailed an appallingly grisly circumstance whose duplication was remarkable. It seems that in both instances the dying person, in 1815 a gentle old lady named Stafford and in 1845 a school-teacher of middle age named Eleazar Durfee, became transfigured in a horrible way; glaring glassily and attempting to bite the throat of the attending physician. Even more puzzling, though, was the final case which put an end to the renting of the house – a series of anaemia deaths preceded by progressive madnesses wherein the patient would craftily attempt the lives of his relatives by incisions in the neck or wrists.
This was in 1860 and 1861, when my uncle had just begun his medical practice; and before leaving for the front he heard much of it from his elder professional colleagues. The really inexplicable thing was the way in which the victims – ignorant people, for the ill-smelling and widely shunned house could now be rented to no others – would babble maledictions in French, a language they could not possibly have studied to any extent. It made one think of poor Rhoby Harris nearly a century before, and so moved my uncle that he commenced collecting historical data on the house after listening, some time subsequent to his return from the war, to the first-hand account of Drs. Chase and Whitmarsh. Indeed, I could see that my uncle had thought deeply on the subject, and that he was glad of my own interest – an open-minded and sympathetic interest which enabled him to discuss with me matters at which others would merely have laughed. His fancy had not gone so far as mine, but he felt that the place was rare in its imaginative potentialities, and worthy of note as an inspiration in the field of the grotesque and macabre.
For my part, I was disposed to take the whole subject with profound seriousness, and began at once not only to review the evidence, but to accumulate as much as I could. I talked with the elderly Archer Harris, then owner of the house, many times before his death in 1916; and obtained from him and his still surviving maiden sister Alice an authentic corroboration of all the family data my uncle had collected. When, however, I asked them what connection with France or its language the house could have, they confessed themselves as frankly baffled and ignorant as I. Archer knew nothing, and all that Miss Harris could say was that an old allusion her grandfather, Dutee Harris, had heard of might have shed a little light. The old seaman, who had survived his son Welcome's death in battle by two years, had not himself known the legend; but recalled that his earliest nurse, the ancient Maria Robbins, seemed darkly aware of something that might have lent a weird significance to the French ravings of Rhoby Harris, which she had so often heard during the last days of that hapless woman. Maria had been at the shunned house from 1769 till the removal of the family in 1783, and had seen Mercy Dexter die. Once she hinted to the child Dutee of a somewhat peculiar circumstance in Mercy's last moments, but he had soon forgotten all about it save that it was something peculiar. The granddaughter, moreover, recalled even this much with difficulty. She and her brother were not so much interested in the house as was Archer's son Carrington, the present owner, with whom I talked after my experience.
Having exhausted the Harris family of all the information it could furnish, I turned my attention to early town records and deeds with a zeal more penetrating than that which my uncle had occasionally shown in the same work. What I wished was a comprehensive history of the site from its very settlement in 1636 – or even before, if any Narragansett Indian legend could be unearthed to supply the data. I found, at the start, that the land had been part of a long strip of the lot granted originally to John Throckmorton; one of many similar strips beginning at the Town Street beside the river and extending up over the hill to a line roughly corresponding with the modern Hope Street . The Throckmorton lot had later, of course, been much subdivided; and I became very assiduous in tracing that section through which Back or Benefit Street was later run. It had, a rumour indeed said, been the Throckmorton graveyard; but as I examined the records more carefully, I found that the graves had all been transferred at an early date to the North Burial Ground on the Pawtucket West Road .
Then suddenly I came – by a rare piece of chance, since it was not in the main body of records and might easily have been missed – upon something which aroused my keenest eagerness, fitting in as it did with several of the queerest phases of the affair. It was the record of a lease in 1697, of a small tract of ground to an Etienne Roulet and wife. At last the French element had appeared – that, and another deeper element of horror which the name conjured up from the darkest recesses of my weird and heterogeneous reading – and I feverishly studied the platting of the locality as it had been before the cutting through and partial straightening of Back Street between 1747 and 1758. I found what I had half expected, that where the shunned house now stood, the Roulets had laid out their graveyard behind a one-story and attic cottage, and that no record of any transfer of graves existed. The document, indeed, ended in much confusion; and I was forced to ransack both the Rhode Island Historical Society and Shepley Library before I could find a local door which the name of Etienne Roulet would unlock. In the end I did find something; some thing of such vague but monstrous import that I set about at once to examine the cellar of the shunned house itself with a new and excited minuteness.
The Roulets, it seemed, had come in 1696 from East Greenwich, down the west shore of Narragansett Bay . They were Huguenots from Caude, and had encountered much opposition before the Providence selectmen allowed them to settle in the town. Unpopularity had dogged them in East Greenwich , whither they had come in 1686, after the revocation of the Edict of Nantes, and rumour said that the cause of dislike extended beyond mere racial and national prejudice, or the land disputes which involved other French settlers with the English in rivalries which not even Governor Andros could quell. But their ardent Protestantism – too ardent, some whispered – and their evident distress when virtually driven from the village had been granted a haven; and the swarthy Etienne Roulet, less apt at agriculture than at reading queer books and drawing queer diagrams, was given a clerical post in the warehouse at Pardon Tillinghast's wharf, far south in Town Street . There had, however, been a riot of some sort later on – perhaps forty years later, after old Roulet's death – and no one seemed to hear of the family after that.
For a century and more, it appeared, the Roulets had been well remembered and frequently discussed as vivid incidents in the quiet life of a New England seaport. Etienne's son Paul, a surly fellow whose erratic conduct had probably provoked the riot which wiped out the family, was particularly a source of speculation; and though Providence never shared the witchcraft panics of her Puritan neighbours, it was freely intimated by old wives that his prayers were neither uttered at the proper time nor directed toward the proper object. All this had undoubtedly formed the basis of the legend known by old Maria Robbins. What relation it had to the French ravings of Rhoby Harris and other inhabitants of the shunned house, imagination or future discovery alone could determine. I wondered how many of those who had known the legends realized that additional link with the terrible which my wider reading had given me; that ominous item in the annals of morbid horror which tells of the creature Jacques Roulet, of Caude , who in 1598 was condemned to death as a daemoniac but afterward saved from the stake by the Paris parliament and shut in a madhouse. He had been found covered with blood and shreds of flesh in a wood, shortly after the killing and rending of a boy by a pair of wolves. One wolf was seen to lope away unhurt. Surely a pretty hearthside tale, with a queer significance as to name and place; but I decided that the Providence gossips could not have generally known of it. Had they known, the coincidence of names would have brought some drastic and frightened action – indeed, might not its limited whispering have precipitated the final riot which erased the Roulets from the town?