How Hierarchical Classification Became the Flat-Earth Theory
Beyond the Flat Earth: Towards an Ontological 'Spherical' Paradigm for Information & Data Governance
Beyond the Flat Earth: Towards an Ontological Paradigm for Information Governance
I recently posted a thought piece on LinkedIn on, “Why hierarchy is no longer fit for the digital age”, and one of the respondents brought up the the idea of being a “flat-earther'“, in the context of ontology, taxonomy, hierarchy and relationality, and linear lifecycle, versus multi-directional continuum thinking. Essentially, the difference between the unidirectional flat hierarchical model versus the multi-dimensional, ‘spherical’ conceptual thinking of ontological approach It made me realise that my central thesis, could be characterised as: designing the real world in hierarchical and taxonomical models is the equivalent of flat-Earther view of the world, and the fully relational and ontological views are the equivalent of a spherical earth, i.e. that a spherical view of the world can accommodate there being parts of the earth that are flat, but a flat earth view cannot accommodate a view where there earth is spherical.
So, this is not to deny the existence of flat earth, (just as I don’t deny that hierarchy exists) just that it is not a model to view or model the earth as a whole. Therefore this is the same of information and data modelling and even how organisations work, i.e. just because there are hierarchical parts, it doesn't mean that the they the whole arct is flat, it can be broadly spherical, but have parts that are flat.
A spherical Earth was not a modern invention; it was a well-established scientific reality in the Hellenistic period. Scholars of the time leveraged compelling empirical evidence to support this understanding. Observations of lunar eclipses consistently showed a round shadow cast by the Earth on the Moon, a shape that would be impossible with a flat disk. Furthermore, the way ships gradually disappear hull-first over the horizon and the change in visible constellations as one travels north or south provided irrefutable proof of the Earth's curvature. This knowledge culminated in remarkable scientific feats, such as Eratosthenes's calculation of the Earth's circumference around 240 BC, a testament to the accuracy of the spherical model. This chimes with the ancient view of Socrates and Plato, that knowledge and information should be expressed and understood ontologically, conceptually and relationally.
Yet, in certain periods of the Middle Ages, this complex, accurate understanding was challenged and at times supplanted by the simpler, more orthodox belief in a flat Earth. This historical regression from a spherical to a flat understanding of the world, often met with resistance and even persecution as seen in the cases of figures like Galileo and Giordano Bruno, provides a powerful metaphor for the current state of information and data governance. Just as Linnaeus denied the history of ontological thought to establish his hierarchical worldview of taxon and phylum, he denied the history of a spherical view to establish his flat view of the Earth. So, while the digital age is fundamentally spherical, a multidimensional, interconnected web, the models we often use to manage it remain stubbornly flat, rooted in hierarchical and taxonomical paradigms that fail to capture the true complexity of our information ecosystems.
The obsolescence of strictly hierarchical models for information and data has been a subject of scholarly discourse for decades. As early as the 1980s, archival theorists began to challenge the traditional, paper-paradigm models of records management, which relied on linear, unidirectional lifecycles and rigid, parent-child classification schemas. This critique was not merely a reaction to nascent digital technologies; it was a recognition that even in a pre-internet era, the complexity of modern organizations had already outgrown these simplistic structures.
Beyond the Flat Earth: Towards an Ontological Paradigm for Information Governance
The prevailing wisdom in data and information governance often defaults to a hierarchical and taxonomical modeling paradigm, a view that, while intuitive, is fundamentally limited in its ability to represent the complexity of modern organisations. This article posits that this approach is analogous to a "flat-earth" perspective, a model that is functional for localised navigation but fails to account for the multidimensional, interconnected nature of the global information landscape. The move towards a fully relational, ontological, and continuum-based understanding of information is the "spherical-earth" equivalent, a paradigm shift that can accommodate and contextualize traditional hierarchies without being constrained by them.
The obsolescence of strictly hierarchical models for information and data has been a subject of scholarly discourse for decades. As early as the 1980s, archival theorists began to challenge the traditional, paper-paradigm models of records management, which relied on linear, unidirectional lifecycles and rigid, parent-child classification schemas. This critique was not merely a reaction to nascent digital technologies; it was a recognition that even in a pre-internet era, the complexity of modern organizations had already outgrown these simplistic structures.
In her doctoral thesis, "DESIGNATION BY DESIGN," Karen Trivette cites the prescient work of archival theorists David Bearman and Richard Lytle, who in 1978 argued that the hierarchical approach "is far too simplistic for modern organizations operating in a world of multi-national corporations, inter-governmental units, regulatory organizations, and federal programmes administered by state, provincial, and local governments." Their critique underscores a crucial point: hierarchical schema lack semantic meaning and interoperability across organizational boundaries. Peterson, in 1986, further noted that "some archivists claimed that automation made hierarchies obsolete and unnecessary," foreseeing a future where technology would demand more flexible and interconnected data models.
While organisations may be represented by a hierarchical reporting structure on a chart, the operational reality is a network of highly specialised, often siloed, functional units working in a fully relational and interdependent manner. Information and data ecosystems mirror this reality; they are not strictly a series of parent-child relationships but a complex web of interconnected nodes where data entities relate to each other in multiple and often non-linear ways.
A purely hierarchical model is unable to capture these nuanced relationships, leading to data silos, redundancy, and a fragmented view of the organization's information assets. It forces a complex, multi-faceted reality into a single, simplistic classification, thereby losing critical context and interconnections. The result is an inability to answer complex queries that require traversing multiple dimensions of data, such as understanding all the projects a person has contributed to across various departments and the specific skills they applied, as opposed to simply identifying their direct manager.
The move to a "spherical" or ontological model is not a rejection of hierarchy, but rather a more comprehensive way to understand information. It acknowledges that hierarchies exist and have a valid place within certain contexts, but they are not the sole or primary organising principle. An ontology, as a formal and explicit specification of a conceptualisation, provides a shared vocabulary of entities and their relationships, enabling a multidirectional and holistic understanding of data. This approach shifts the focus from "where does this data belong in a tree?" to "how is this data related to everything else?"
By embracing a relational and continuum-based approach, organisations can move beyond the limitations of linear lifecycle models and rigid classification systems. Information governance becomes less about enforcing a single, top-down hierarchy and more about fostering a shared understanding of interconnected data. This paradigm shift, from a "flat-earth" view of data as a series of isolated hierarchies to a "spherical" view of data as a dynamic, interconnected knowledge graph is essential for navigating the complexity of the digital age and unlocking the true value of information assets. It is a conclusion that archival theorists, from Bearman and Lytle to Trivette, have been advocating for decades, a testament to its enduring relevance far beyond the confines of the internet era.