The Metadata Mirage

This post is part of a series that explores how metadata can be used as a site of resistance, refusal, and poetic subversion. From classification to linked data, the series investigates how cataloging practices can encode oppression, and how they can be reimagined to challenge dominant systems and speak from the margins. Check all the posts in this section's index.

Metadata Is Not Just Description

Metadata is often defined as "data about data," a phrase so deceptively concise that it masks the density of what metadata actually does. Behind its minimalist framing lies an expansive apparatus: schemas, vocabularies, controlled values, ontological hierarchies. These are not neutral tools. They do not simply store information — they impose structure. And structure, in any knowledge system, is never innocent.

To describe something is to decide how it will be known. Metadata doesn't just point to a document or object — it interprets it, encodes assumptions about its nature, and determines its position within a larger framework of meaning. That framework is rarely questioned. Yet it is precisely in the structure of metadata —in what it asks for, in what it makes optional, in what it does not permit— that entire epistemologies are either validated or suppressed.

Consider a simple example: the field dc:language. On the surface, it appears harmless. But it typically accepts ISO 639 codes — a standard that recognizes certain languages and ignores thousands of others, particularly oral, unwritten, sacred, or community-bound varieties. A language without an ISO code becomes illegible to the system. And what is illegible often becomes unpreservable, unsearchable, unacknowledged.

The same applies to fields like dc:type, dc:subject, or dc:coverage. These are not mere containers. They are editorial decisions in code. And each one, by what it does and does not allow, participates in the construction of an institutional worldview.

What the Field Hides

One of metadata's most dangerous qualities is its ability to render its exclusions invisible. Because metadata is framed as technical rather than ideological, its silences are mistaken for neutrality. A blank field is seen as a gap in the data — not as the result of a schema that refuses to accommodate a particular kind of knowledge.

Yet these omissions are everywhere.

An ancestral narrative is recorded as a "story." A ceremonial object is listed as "material culture." A prayer-song is indexed under "folklore." A document written in Quechua but transmitted orally is marked "no linguistic content." These aren't accidents — they're the product of fields that do not recognize relational, embodied, or plural knowledge forms. The schema does not know how to hold them, and so it doesn't. They are misrepresented, distorted, or made to fit within existing categories — often categories designed with very different ontological premises.

When metadata becomes institutional infrastructure —inside catalogs, digital repositories, archival records, data portals— its exclusions become encoded into access, retrieval, and memory. What cannot be named within the schema often cannot be found. What cannot be found cannot be cited, referenced, reused, or acknowledged. Over time, these exclusions accumulate into structural invisibility.

The Mirage of Neutrality

The central illusion of metadata is that it merely describes. That it is a passive, impartial layer floating above the content. This illusion is what enables metadata to function as a disciplinary device without being recognized as such.

Metadata does not sit above knowledge. It sits inside it — shaping what counts as valid, what counts as finished, what counts as knowable. When we describe a knowledge object using a predefined set of fields and controlled terms, we do not just store it for future retrieval. We reconstitute it according to the logic of the system. We reduce its complexity in the name of compatibility. We erase its ambiguity in the name of legibility. We force it to travel through a syntax not of its own making.

This violence is particularly acute when dealing with non-Western or non-textual knowledge systems. Metadata standards overwhelmingly reflect Eurocentric, document-centered, object-oriented models of information. They assume that knowledge can be discretized, named, and made portable. They assume that description is always permissible — that all knowledge can, and should, be encoded.

But some forms of knowledge are intentionally withheld. Some are seasonal, contextual, communal, or sacred. Some resist fixing. Others demand relational context to exist at all. The question is not whether metadata can adapt to these realities — but whether it is willing to acknowledge that it was never designed for them in the first place.

To Structure Otherwise

To work with metadata in this context is not just a technical act — it is a political one. It means confronting the authority of standards, questioning the ideological premises of schemas, and recognizing the asymmetries they perpetuate. It means understanding that every metadata field encodes an epistemological choice, and that these choices have consequences: not just for search and retrieval, but for memory, representation, and legitimacy.

It also means refusing the comfort of minimal compliance. A field like dc:description might be open-ended, but that openness is not enough. We must ask whether its openness can hold multiplicity, or whether it simply absorbs difference into a neutralized field. We must ask whether structured data can coexist with ambiguity — or whether it always seeks to resolve it.

This refusal is not an abstract critique. It is a practical stance. It is the decision to treat each record as a site of negotiation. To ask what is missing, what is misnamed, what is made invisible. It is the choice to foreground epistemic responsibility, even —especially— when the system makes it easy not to.

Where the Work Begins

This is not a rejection of metadata, but of its uncritical deployment. It is an insistence that description is never just description — that structure is always situated, and that what we call metadata is often a misrecognized expression of institutional worldview. When we treat metadata as innocent, we relinquish our role as stewards of epistemic integrity. When we recognize its politics, we can begin to use it differently — carefully, reflexively, and, when needed, against itself.

That is where the work begins: with the clarity that metadata is never neutral, and the responsibility to act accordingly.