Pura Belpr\u00e9, the first Puerto Rican librarian hired by the New York Public Library, built something that was simultaneously a collection, a community, and a cultural argument. Her work brought Spanish-language materials into the library, creating programming that reflected the experiences of New York’s Puerto Rican community, insisting that the library’s collections should represent the full range of its community’s cultural life.\u00a0 She believed collection development as social design. She understood that a collection is not just a set of objects. It is a set of relationships.\u00a0 It is between the institution and its community members, between the materials and their users, between the past that the collection preserves and the future that it enables.<\/p>\n
This is the deepest lesson of the greats: that a collection is not a thing. <\/strong><\/p>\n The designed collection is not the product of a single visionary decision. <\/strong><\/p>\n It is the product of thousands of decisions, made over decades, by people who understood that they were building something larger than themselves \u2014 something that would outlast them, serve people they would never meet, and preserve knowledge that would otherwise be lost. That understanding is what separates the great collections from the merely large ones. It is what the greats teach us, and it is what the next generation of collection builders must learn.<\/p>\n Every collection is defined as much by what it does not contain as by what it does. The decision not to collect is as consequential as the decision to collect, and it requires the same quality of long-horizon thinking.<\/p>\n The great special collections and archives of the world are defined by their scope and by the decisions made by professionals over decades.\u00a0 It is about what falls within the collection’s purview and what does not. The Folger Shakespeare Library in Washington, D.C., holds the world’s largest collection of Shakespeare’s printed works not because it collected everything, but because its founders, Henry Clay Folger and his wife Emily Jordan Folger, made a specific, sustained, and disciplined decision to focus on a defined scope and to pursue that scope with extraordinary depth. The Folger’s collection of Shakespeare’s First Folio.\u00a0 There is 82 of the approximately 235 known surviving copies there.\u00a0 It is the result of a collection development philosophy that prioritized depth over breadth, that understood the value of concentration, and that was willing to pay prices that seemed extravagant at the time in pursuit of a collection that would be irreplaceable.<\/p>\n This kind of disciplined scope definition is the foundation of any great special collection, and it is the aspect of collection development that is most frequently sacrificed to short-term institutional pressures. The pressure to accept donations and to say yes to the collection of papers that does not quite fit the scope but comes with a significant gift.\u00a0 This is one of the most persistent threats to research collections managers. Every out-of-scope donation accepted is a processing burden, a storage burden, a space and a descriptive burden that diverts resources from the collection’s core mission. It is also a signal to future donors about what the collection will accept, and that can turn off investments in research.<\/p>\n Archibald MacLeish, who served as the 9th Librarian of Congress from 1939 to 1944 and is one of the most consequential figures in the Library’s institutional history, understood collection development as a form of intellectual argument. His reorganization of the Library \u2014 which transformed it from a relatively passive repository into an active participant in American intellectual and cultural life.\u00a0 His clear articulation of what the Library of Congress was for and what it should therefore collect was historic.<\/p>\n MacLeish’s vision was expansive: the Library should be the memory of the American people, and its collections should reflect the full range of American experience and American thought. But expansiveness of vision does not mean absence of principle. MacLeish’s collection development philosophy was grounded in a clear understanding of the Library’s unique position.\u00a0 It has legal deposit privileges,\u00a0 and its relationship to the federal government, its role as a resource for Congress and for the nation is a challenge.\u00a0 It is with that understanding that everyone who works at the LOC shapes every collection decision made at some time during their tenure.<\/p>\n The relationship between collection development and the other three domains:<\/p>\n A collection that cannot be found is not a collection. It is a warehouse.<\/strong><\/p>\n This is the insight that drives the work of cataloging and description, and it is an insight that is perpetually under pressure from institutional forces that treat cataloging as a cost center rather than a core function. The history of great collections is, in significant part, a history of institutions that resisted this pressure \u2014 and the history of failed collections is, in significant part, a history of institutions that did not.<\/p>\n Charles Ammi Cutter, whose Rules for a Dictionary Catalog published in 1876 established the theoretical foundations of modern library cataloging, articulated the purpose of the catalog with a precision that has not been improved upon: the catalog should enable a user to find a book when the author, title, or subject is known; to show what the library has by a given author, on a given subject, or in a given kind of literature; and to assist in the choice of a book as to its edition or as to its character. These three functions \u2014 finding, collocating, and selecting \u2014 remain the fundamental purposes of any descriptive system, whether it is a card catalog, a MARC record, an EAD finding aid, or a linked data graph.<\/p>\n What Cutter understood, and what the best cataloging theorists since him have continued to develop, is that description is not neutral. Every cataloging decision \u2014 what to call a subject, how to structure an authority record, what level of description to apply to a collection \u2014 encodes assumptions about who will use the collection and what they will be looking for. Sanford Berman, who spent decades as head cataloger at the Hennepin County Library in Minnesota, made this point with characteristic directness: the Library of Congress subject headings that dominated American library catalogs for most of the 20th century were built on assumptions about who library users were and what they cared about that systematically excluded large portions of the population. His campaign to reform those headings \u2014 to replace terms like “Yellow Peril” and “Jewish question” with language that reflected the perspectives of the communities being described \u2014 was not merely a matter of political correctness. It was a matter of access. A subject heading that a researcher cannot recognize as relevant to their inquiry is a subject heading that fails its fundamental purpose.<\/p>\n For special collections and archives, the descriptive challenge is compounded by the nature of the materials. Unlike published books, which arrive with title pages and established bibliographic identities, archival collections are unique, often unprocessed, and frequently described at the collection level rather than the item level. The finding aid \u2014 the document that describes the contents, arrangement, and context of an archival collection \u2014 is both a navigational tool and an interpretive act. A well-constructed finding aid does not merely list what is in a box. It explains why the collection exists, who created it, what its relationship is to other collections, and what a researcher might find in it that they did not know to look for.<\/p>\n Encoded Archival Description (EAD), the XML-based standard that brought finding aids into the digital environment, was built on the recognition that archival description has a structure \u2014 that the relationship between a collection and its series, between a series and its folders, between a folder and its items, is a hierarchical relationship that can be encoded and made machine-readable. This encoding makes finding aids searchable, aggregable, and interoperable across institutions. A researcher can now search across the finding aids of hundreds of institutions simultaneously, identifying collections relevant to their work that they would never have found through a single institution’s catalog.<\/p>\n But the power of this infrastructure depends entirely on the quality of the description that feeds it. A finding aid that describes a collection as “papers relating to various subjects, 1920-1965” is not a finding aid. It is a placeholder. The investment in descriptive infrastructure \u2014 in the training of archivists, in the time allocated for processing, in the standards and authorities that ensure consistency across descriptions \u2014 is an investment in the discoverability of the collection, and therefore in its usefulness, and therefore in its institutional justification.<\/p>\n Carla Hayden, the 14th Librarian of Congress and the first woman and first African American to hold that position, has made the expansion of digital access to the Library’s collections a central priority of her tenure. Her instinct \u2014 that a collection’s value is realized only when it is accessible, and that accessibility in the 21st century requires digital infrastructure \u2014 is a direct extension of Cutter’s insight into the contemporary environment. The cataloging and description work that makes a collection discoverable is not separate from the collection development work that builds it. They are the same work, understood at different scales.<\/p>\n Henriette Avram, whose development of the MARC (Machine-Readable Cataloging) standards in the 1960s transformed the descriptive infrastructure of libraries worldwide, was not a preservation specialist. But her work illustrates a principle that applies directly to environmental design: standards must be built to accommodate the full range of what they will be asked to describe.<\/p>\n MARC was designed not just for books but for the full range of library materials \u2014 maps, music, visual materials, computer files \u2014 because Avram understood that a descriptive standard that could not accommodate the collection’s actual diversity would eventually become a constraint on the collection’s development.<\/p>\n The same principle applies to environmental infrastructure. A climate control system designed only for paper will become a constraint \u2014 and eventually a source of damage \u2014 when the collection grows to include photographic materials, magnetic media, or objects. The long-horizon thinker designs environmental systems for the collection that will exist in twenty years, not the collection that exists today.<\/p>\n Thomas Bodley, who refounded the Bodleian Library at Oxford in 1598 after its earlier collections had been dispersed, understood this in architectural terms. His insistence on a dedicated library building \u2014 separate from the university’s other functions, designed specifically for the storage and use of books \u2014 was not merely a matter of institutional prestige. It was a recognition that the physical environment of a collection must be purpose-built, not adapted from spaces designed for other uses.<\/p>\n The Duke Humfrey’s Library that Bodley restored and expanded, with its chained books and its carefully controlled access, was an environmental system as much as it was a reading room. The chains were not security theater. They were a response to the specific vulnerability of the collection \u2014 the risk that books would be removed and not returned \u2014 just as a modern cold vault is a response to the specific vulnerability of photographic materials.<\/p>\n The building is not a container for the collection. The building is an argument about what the collection is for.<\/p>\n This is the insight that separates the great library architects from the competent ones, and it is an insight that requires the architect and the librarian to work in genuine collaboration \u2014 not in the sequential relationship where the librarian specifies requirements and the architect designs to them, but in the iterative, mutually educating relationship where each party’s understanding of the problem is transformed by the other’s expertise.<\/p>\n The Beinecke Rare Book and Manuscript Library at Yale, designed by Gordon Bunshaft and opened in 1963, is the most discussed example of library architecture as collection argument in the 20th century. Its defining feature \u2014 the translucent marble panels that filter natural light into the central book tower \u2014 is not a decorative choice. It is a preservation decision. The marble transmits diffused light while blocking ultraviolet radiation, creating an environment in which the collection is visible without being damaged by the light that makes it visible. The building solves a fundamental tension in special collections architecture \u2014 the tension between display and protection \u2014 through a material choice that is simultaneously structural, environmental, and symbolic.<\/p>\n The book tower at the center of the Beinecke is not accessible to readers. It is a vault, visible through glass walls from the reading room that surrounds it. This spatial arrangement encodes a specific argument about the relationship between the collection and its users: the collection is present, it is visible, it is the reason for the building’s existence, but it is not directly accessible. Access is mediated \u2014 through the reading room, through the request process, through the staff who retrieve materials and deliver them to supervised tables. This mediation is not bureaucratic obstruction. It is preservation policy expressed in architectural form.<\/p>\n Compare this to the spatial logic of a public research library like the New York Public Library’s Stephen A. Schwarzman Building on Fifth Avenue, where the Rose Main Reading Room \u2014 297 feet long, 78 feet wide, 52 feet high \u2014 is designed to accommodate large numbers of researchers working simultaneously in a shared space. The architecture here makes a different argument: that research is a public activity, that the library is a civic institution, that the scale and grandeur of the space communicate the importance of the intellectual work conducted within it. The collection is not visible from the reading room in the way it is at the Beinecke; it is held in stacks below the building, retrieved by pneumatic tube and dumbwaiter. But the spatial experience of the reading room itself is part of the collection’s value proposition \u2014 it is the environment in which the collection is used, and it shapes the quality and character of that use.<\/p>\n John Cotton Dana, who transformed the Newark Public Library in the early 20th century and is often credited with inventing the concept of the active, community-engaged public library, understood spatial design as collection policy in a different register. His insistence on open shelves \u2014 on allowing patrons to browse the stacks directly rather than requesting materials through a counter \u2014 was a spatial argument about the relationship between the library and its community. Open shelves communicate trust. They communicate that the library’s purpose is use, not preservation. They make serendipitous discovery possible in a way that closed stacks cannot. Dana’s spatial choices encoded a philosophy of access that was inseparable from his collection development philosophy.<\/p>\n For contemporary special collections and archives, the spatial design challenge is compounded by the need to accommodate both analog and digital workflows. A processing room that was designed for the arrangement and description of paper-based collections must now accommodate the digitization equipment, the networked workstations, and the storage infrastructure required to create and manage digital surrogates. A reading room that was designed for researchers consulting physical materials must now accommodate researchers who are working with digital files on their own laptops while simultaneously consulting physical materials at the table. These are not trivial spatial requirements. They require electrical infrastructure, network infrastructure, furniture design, and lighting design that were not part of the vocabulary of library architecture a generation ago.<\/p>\n The long-horizon spatial thinker designs for flexibility \u2014 for the ability to reconfigure spaces as workflows change, as collections grow, as technologies evolve. This means avoiding the temptation of highly specialized spaces that serve current needs perfectly but cannot be adapted to future needs at all. It means building in excess capacity \u2014 in electrical load, in network bandwidth, in floor load ratings for compact shelving \u2014 that will seem wasteful in the short term and will prove essential in the long term.<\/p>\n Begin with a failure, because failures help us reflect and learn best practices.<\/p>\n In the mid-20th century, a number of American research libraries made aggressive decisions to expand their collections through microfilm. The logic was logical in the short term: microfilm was stable, compact, and reproducible. It solved the space problem and the preservation problem simultaneously. In fact, we were hired by institutions to investigate strategies to sell the film, because of outdated and mismatched tech.<\/p>\n What they did not invest in the book?\u00a0 What was it about microfilm that made it better?\u00a0 It saved space, but it really required specific environmental conditions to remain stable: low temperature, low humidity, controlled light exposure or it will become chipped and unusable. Many institutions stored their film and ask to remove it during library planning projects.\u00a0 It is appropriate for paper to be part of the library, but the acetate base of the film is outdated. For example, the vinegar syndrome (named for the acetic acid released as acetate degrades) renders millions of frames unreadable today. Lastly, the originals were purchased as part of an investment group and we tried to sell it.\u00a0 There was no interest and it had been discarded.<\/p>\n This is not a story about technology failure. It is a story about integrated system failure. The collection development decision (acquiring microfilm) was made because it could save space.\u00a0 \u00a0Without adequate understanding of the preservation required it failed as a collection development strategy.\u00a0 Of course, today we have digital files that serve the same purpose.\u00a0 They save space.<\/p>\n S.R. Ranganathan, the Indian mathematician and librarian whose Five Laws of Library Science remain the most compressed and durable statement of library purpose ever written, understood this integration intuitively.<\/p>\n His fifth law \u2014 “the library is a growing organism<\/strong>” \u2014 is often quoted as a celebration of collection growth. It is more usefully read as a warning. Organisms that grow without integrated systems \u2014 without circulatory systems, without nervous systems, without metabolic regulation \u2014 do not thrive.<\/p>\n They metastasize. The growing library that does not grow its environmental infrastructure, its descriptive infrastructure, and its spatial infrastructure in parallel is not fulfilling Ranganathan’s fifth law. It is violating it.<\/p>\n The long horizon, then, is not simply a matter of planning further into the future. It is a matter of planning across more domains simultaneously, and understanding that decisions made in any one domain will propagate consequences \u2014 sometimes decades later \u2014 into all the others.<\/p>\n Antonio Panizzi, who served as chief librarian of the British Museum library from 1856 and is responsible for the magnificent Round Reading Room that defined the intellectual life of Victorian London, understood something that many of his contemporaries did not: the physical environment of a collection is not a background condition. It is an active participant in the collection’s survival or destruction.<\/p>\n Panizzi fought for adequate space, for proper ventilation, for conditions that would slow the deterioration of the Museum’s holdings. He was not a preservation scientist \u2014 the field did not yet exist in its modern form \u2014 but he had the instinct of a systems thinker. He understood that a book that cannot be read because it has crumbled is not a book. It is a record of a failure of institutional responsibility.<\/p>\n Modern preservation science has given us the precision that Panizzi could only approximate through instinct. We know now, with considerable specificity, what different materials require:<\/p>\n Paper-based materials<\/strong>\u00a0\u2014 the bulk of most research collections \u2014 are best preserved at temperatures between 60 and 70 degrees Fahrenheit, with relative humidity between 30 and 50 percent. Fluctuation is as damaging as absolute levels; a collection that cycles between 40 and 70 percent relative humidity will deteriorate faster than one held steadily at 55 percent. Light exposure, particularly ultraviolet light, accelerates the degradation of paper and ink. Acid migration from poor-quality housing materials compounds the damage.<\/p>\n Photographic materials<\/strong>\u00a0require differentiated treatment. Black-and-white silver gelatin prints and negatives are relatively stable under standard cool storage conditions. Color photographs, which rely on organic dye layers that fade at different rates, require colder storage \u2014 often below 35 degrees Fahrenheit for long-term retention. Motion picture film, particularly nitrate-base film produced before 1951, is not merely unstable but actively dangerous: nitrate film is flammable and must be stored in dedicated vaults with fire suppression systems designed specifically for its properties.<\/p>\n Magnetic media<\/strong>\u00a0\u2014 audiotape, videotape, floppy disks \u2014 present a different set of challenges. The binder that holds the magnetic particles to the tape substrate is susceptible to hydrolysis, a process in which moisture causes the binder to break down, making the tape sticky and unplayable. Cold, dry storage slows this process. But cold storage also requires that media be allowed to acclimate to room temperature before playback, a step that is frequently skipped in operational environments and that causes significant damage.<\/p>\n Parchment and vellum<\/strong>, the animal-skin materials on which medieval manuscripts were written, are hygroscopic \u2014 they absorb and release moisture from the surrounding air, expanding and contracting as they do. A manuscript that has been stable for five centuries in a European monastery at consistent humidity will begin to cockle and distort if moved to a modern climate-controlled environment where the humidity is held at a different level. The transition itself is the danger. Preservation of these materials requires not just the right conditions but a managed transition to those conditions.<\/p>\n The implications for spatial design are direct and non-negotiable. A collection that holds all of these material types \u2014 as any serious research collection will \u2014 cannot be housed in a single undifferentiated space. It requires zones: a general stack environment for paper-based materials, a cold vault for photographic and audiovisual materials, a dedicated space for nitrate film if the collection holds it, a climate-stable reading room where materials can be used without being subjected to the fluctuations of a public building’s HVAC system.<\/p>\n The argument of this post can be stated simply: the four domains of collection building:<\/p>\n They are components of a single integrated system, and the failure to understand them as such is the most reliable predictor of collection failure.<\/strong><\/p>\n The great collection builders understood this integration intuitively or learned it through experience. Panizzi understood that space and environment were inseparable from collection value. Cutter understood that description was inseparable from access. Bodley understood that architecture was inseparable from institutional mission. MacLeish understood that collection development was inseparable from institutional identity. Avram understood that descriptive standards were inseparable from the full range of materials they would be asked to describe.<\/p>\n What these figures shared was not genius in any single domain. It was the capacity for systems thinking in our work.\u00a0 It is the ability to hold multiple domains in mind simultaneously and to understand that a decision made in one domain would propagate consequences into all the others. This capacity is not innate. It is developed through practice, through collaboration, through the willingness to learn enough about adjacent domains to understand their constraints and their possibilities.<\/p>\n For the contemporary practitioner, this means that the archivist who does not understand the basics of environmental science is not fully equipped to make collection development decisions. The architect who does not understand the basics of archival processing is not fully equipped to design a special collections facility. The cataloger who does not understand the basics of collection scope is not fully equipped to make descriptive decisions. And the administrator who does not understand all four domains is not fully equipped to lead an institution that aspires to build a great collection.<\/p>\n The long horizon is not a library planning horizon. It is a thinking horizon for library development.\u00a0 It is the capacity to define the risks of today’s decisions across time and across domains, and to understand that the collection being built now will be used, maintained, described, and housed by people who do not yet exist, in circumstances that cannot be fully anticipated, and that the decisions made today will either enable or constrain their work.<\/p>\n","protected":false},"excerpt":{"rendered":" Lessons from the Greats on Building Archives, Special Collections, and Research Libraries An Essay for Advanced Knowledge Management Practitioners What the Greats Teach Us Pura Belpr\u00e9, the first Puerto Rican…<\/p>\n","protected":false},"author":6,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[1725,54,10,1724],"class_list":["post-8941","post","type-post","status-publish","format-standard","hentry","category-library-planning-research","tag-history-of-library-science","tag-library-consultant","tag-library-design","tag-library-programming-and-design"],"yoast_head":"\n\n
Collection Development: The Architecture of Scope<\/h2>\n
Folger Shakespeare Library<\/h3>\n
Collection Development is a Form of Intellectual Argument<\/h3>\n
\n
The Architecture of Access<\/h2>\n
LESSONS LEARNED FROM CUTTER<\/h2>\n
XML-based standards<\/h3>\n
A collection’s value is realized only when it is accessible<\/h3>\n
MARC STANDARDS<\/h2>\n
Spatial Architecture: The Building as Collection Policy<\/h2>\n
Light is a Filter to Knowledge<\/h3>\n
The Book Tower<\/h3>\n
John Cotton Dana<\/h3>\n
\nWhy Collections Fail When Thinking Is Short<\/h2>\n
\nEnvironmental Control and the Architecture of Survival<\/h2>\n
\nSystem Thinking That Makes Collections Great<\/h2>\n
\n