Eli Hill Case Study I:  The Asylum in the Archive
Eli Hill Case Study I: The Asylum in the Archive

Eli Hill Case Study I: The Asylum in the Archive

Cases from the Asylum Archive

Eli Hill (1832-1877)

Robert C. Allen

In getting my books, I have been always solicitous of an ample margin; this not so much through any love of the thing itself, however agreeable, as for the facility it affords me of pencilling suggested thoughts, agreements, and differences of opinion, or brief critical comments in general.  Where what I have to note is too much to be included within the narrow limits of a margin, I commit it to a slip of paper, and deposit it between the leaves; taking care to secure it by an imperceptible portion of gum tragacanth paste.

                                –-Edgar Allan Poe, “Marginalia,” Democratic Review, Nov. 1844

imperceptible portion of gum tragacan imperceptible portion of gum tragacan

The Asylum in the Archive

Since 2017, UNC’s Community Histories Workshop (CHW) has been excavating historical records relating to insanity and mental health in North Carolina, prompted by our discovery of comprehensive admissions ledgers from the Dorothea Dix Hospital in Raleigh, North Carolina, covering the period from the hospital’s founding in 1856 to 1920.  A change in the state’s open records law in 2016 made publicly accessible state records created more than 100 year ago.  The admissions ledgers and other records from this period relating to the operation of the state’s first and, for many years, principal insane asylum are available to all who wish to examine and copy them in the Reading Room of the State Archives of North Carolina.

Sarah Almond and Lucas Kelley, photographing admissions ledgers in the Reading Room of the State Archives of North Carolina in 2018

In the fall of 2017, the CHW made a commitment to transcribe the 7000+ individual patient records from the admissions ledgers starting with patient #1 in February 1865 up to the beginning of the rolling 100-year sequester period.  CHW Assistant Director, Sarah Almond, devised a transcription template that allowed a team of specially trained undergraduate research fellows (Dani Callahan, Hannah Frisch, Thomas Burnett, and Keely Curry) to transform handwritten nineteenth-century notations into structured data, which, in turn, became the basis for the first comprehensive, searchable database of a 19th century U.S. insane asylum.

The admissions ledgers are essentially 19th century spreadsheets: each row a patient admission and each column an attribute of that patient, including name, marital status, age, and the “supposed cause” and “form” (diagnosis) of the patient’s insanity.

Each entry continues on a facing page, providing information on the patient’s county of residence, length of institutionalization, and eventual disposition and condition: cured, improved, not improved; discharged, transferred, “eloped” (escaped) or died (including cause of death).  In all each record contains some 20 datapoints.

The Dix admissions ledgers are extraordinary resources for documenting the individuals who were treated at North Carolina’s first and primary insane asylum, as well as revealing how “madness” and its various expressions were categorized.  The database created from the ledgers makes it possible to ask questions about relationships among groups of patients both at a single moment in the past and across the hospital’s first 64 years of operation (1856-1920): how did the patient population break down in terms of age, gender, geography, supposed cause, form, and disposition?  How many patients died at the hospital and what were the causes of their deaths (as understood and described at the time)?

We might think of the ledger in dramatic or narrative terms.  Each row establishes a character by name, age, gender, marital status, nativity, and (in the absence of a specific notation) race: Anna M. Kirkland (the first female patient at the hospital) was a 39 year old white woman from Orange County, North Carolina.  That character is then defined in relation to mental state (mania) and the reason for her being brought to the hospital (domestic trouble).  The only information it gives about the patient’s life before that day is the duration of the “attacks”—in Annie’s case, ten years.

Then as we scan left to right across the open ledger we learn what “happened” to this character following her time at the asylum and why: in Annie’s case, she died there some thirty-three years and eight months after the day she entered as a result of influenza and heart failure.

The ascription of “supposed cause” we might think of as the title of another narrative, the protagonist of which is the patient but contents of which we can only guess.  The narrator of that story might have been a family member who accompanied the patient (and might have been the person responsible for initiating the legal process that resulted in her being found “insane”) or the county sheriff’s deputy charged with transporting the patient.  It might have been construed by the hospital staff member (not identified in the ledger) who was responsible for what we would now call “intake” from the commitment papers submitted with the patient.

Some supposed causes would seem to be based on a prior family or social drama involving other (unidentified) characters: “Domestic Trouble,” “Ill Treatment,” Unrequited Love.”  Some point to physical trauma: “Sunstroke,” “Childbirth.” Some attribute insanity to emotional states: “Jealousy,” “Grief,” “Religious Excitement.” Some to loss: of a child, of a husband, of property. What is behind/before the attribution “Masturbation,” “Syphilis,” “Intemperance,” “Puerperal Insanity,” “Opium,” “Disappointment Ambition?” There are more than one hundred different supposed causes listed in the ledgers.  Each suggests then obscures a precipitating narrative.  The narrator leaves the patient and us behind.

When people view the admissions ledgers for the first time, they are often most curious about the story that lies in what printers call the “gutter”: the point at which two facing pages disappear into the binding.  In this case the gutter is temporal: what happened between the moment a patient was admitted to the asylum (on the outer margin of the verso) and the moment (on the outer margin of the recto)—a few months to a few decades later–at which she leaves?

Beginning in 1887, the asylum began to keep more extensive intake records, called general case books, on which there was space for treatment notes.  Sadly, the practice of recording treatment notes continued only sporadically (too many patients?  too few staff?).  We can only extrapolate from treatment records surviving from other asylums and from the articles asylum superintendents published in The American Journal of Insanity.  We know, for example, that when Anna (Cameron) Kirkland was first admitted to the Western Insane Asylum in Staunton, Virginia, in 1846, she was given opium at night and mercury (a powerful emetic in addition to being a poison) in the morning.  However, we know little about how her terrible recurrent melancholy was treated over the more than thirty years she was at Dix.

Throughout our engagement with the asylum records, we have tried to resist (and encouraged others to resist) applying 21st century diagnostic labels to nineteenth century individuals.  However, we do believe there is value in examining how the nomenclature and nosology (clinical categorization) of mental illness were established in the middle decades of the 19th century and how it changed over the next half century, by which time (1920s) the Association of Medical Superintendents of American Institutions for the Insane (AMSAII) had become the American Psychiatric Association.  As both mental health professionals and those in treatment for a specific condition are aware, contemporary psychiatric diagnoses are not stable, unchanging, uncontested  categories.  In the first edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM I) in 1952, homosexuality was classified as “sociopathic personality disturbance.”  The following edition, DSM II (1968), labeled it “sexual deviation.”  By 1973, in the wake of both research findings and protests within and beyond the psychiatric community, homosexuality was dropped from all future editions of the DSM.

There is also the question of the meaning of a given diagnostic term applied in the 19th century and what would seem to be its cognate in the 21st.  For example, dementia is applied hundreds of times in the admissions ledgers, but its use does not correspond to what we think of today as senile dementia: the supposed cause of the first patient so diagnosed in the ledgers (1856) is “blow to the head,” and yet he remained in the asylum for more than fifteen years.

Even when we have shared copies of original records with our colleagues in the Medical School and Department of Psychiatry (including leading “Grand Rounds” for the Dept. of Psychiatry in January 2020), we have encouraged them instead to ask “What more about this person/patient do we most want to know?”  Our strategy has been to (sequentially) combine individual close readings of individual records with social (small group) readings of several different records, to facilitate the generation of questions, recognition of difference, and discovery of connections.

But there is also a compelling historiographic rationale here: we do not want to reproduce the reductive logic of nineteenth-century vital statistics in our own practice.  If we start with pre-existent analytical categories, we run of the danger of reducing individuals to “cases” in a crude epidemiological sense, to an illustrative “micro-history” case where the individual’s role is either to be an instance or an outlier.  Rather, I would urge us to think of the asylums records as, in the first instance, a field in an archaeological sense, but a field without boundaries and without a landmark to serve as an orienting center.  Thus, it doesn’t matter where we start in trying to come to terms with these records; no fixed criteria by which one record/one person deserves our historiographic attention more than another.  There is no Minoan palace, no Roman fort lying at some distance beneath the surface of this field—no structure that will impose meaning and relevance to whatever pottery shards we might find. The answer to the question “Where should we start?” is always “here.” The answer to the question, “Is this important?” is always “yes.”  The records hint at a rhysomatic landscape—one of possible connections, of fungi not oaks.  But the terms of these connections are not “there” prior to the connections drawn between them. They are not so much discovered as provisionally proposed.  Drawn in pencil.

In short, the admissions ledgers give us only the most basic information about the transformation from citizen to inmate, from someone with a spouse, father, mother, children, grandparents, neighbors, homes, and communities to someone stripped of these connections, recast in relation to a psychiatric disorder, and made a part of a new and essentially carceral community: that of the insane asylum.  The information that was noted about each patient at the time of admission, (name, gender, marital status, nativity, etc), the data and circumstances of disposition (discharged, transferred, eloped, died), and the condition of the patient at the time of disposition (cured, improved, unimproved, died) were collected in order to form “vital statistics” regarding the patient population.  The reporting of vital statistics, along with other numerical summaries (cost per patient, value of farm goods produced, etc.) were regularly required of the asylum superintendent, as we can see from the report submitted by the asylum’s first superintendent, Dr. Edward Fisher, at the end of October, 1857.

Nothing in the admissions ledger allows us to restore individuality and personal history to a given patient, to roll back the calendar to the day before they arrived at the asylum’s portico.  Nor can we fast-forward to the day after a patient was discharged or transferred to another institution.  We are not told what effect a patient’s death might have had on their family or, in the case of patients who had been dropped off by family members decades before, if there was any family left to mourn.  This would require that we return the admissions ledgers to the stacks where they are kept and leave the archive Reading Room where we first encountered them.  To mix metaphors: it means that we have to parachute from the 30,000 foot level from which we can make out the categorical contours of all 7200 records and get on our historiographic hands and knees to see if any footprints survive from a single life.

Proceed to next chapter