An archive does not simply accumulate — it accumulates in a way that produces, over time, a kind of resistance to its own use, a friction that increases with size even as the archive's nominal purpose is to make material available. Bennett's account of vibrant matter — her argument that material objects possess a kind of agency, a capacity to act and to affect, that is not reducible to human intention or use — offers an unexpected lens on this problem: an archive's materials do not passively wait to be read, they exert their own pressure on the conditions of reading, their sheer accumulated mass and heterogeneity becoming an active force that shapes — and can exhaust — the attention brought to bear on them. What might be called ArchiveFatigue names this condition: not simply the fatigue of the researcher confronting too much material, but a kind of fatigue intrinsic to the archive itself, a state in which the accumulated material has reached a density at which further accumulation does not increase legibility but begins to work against it, in which each new deposit makes the whole marginally harder to navigate even as it makes the whole marginally more complete. Hui's account of digital objects — his argument that data and metadata exist in a recursive relationship in which each level of organization is itself an object that can become data for a higher level of organization — describes one structural source of ArchiveFatigue: an archive organized recursively can in principle accommodate unlimited growth, but the cost of this accommodation is that navigating the archive requires navigating its own structure of organization, which itself grows and changes, so that familiarity with the archive's contents does not guarantee familiarity with the current state of its organization. What allows an archive to remain usable despite, or through, this accumulating fatigue is not a reduction in size — archives that matter tend not to shrink — but the development of pathways, indexes, and orientations that allow specific kinds of access without requiring confrontation with the whole. This might be called LegibleArchive: not an archive that has become smaller or simpler, but an archive that has developed enough internal differentiation — enough distinct entry points, enough varied indexing logics, enough redundant pathways to the same material — that different users with different needs can each find a route through that does not require them to process the archive's total fatigue at once. Flusser's account of technical images — his argument that the photograph and its successors are not windows onto the world but surfaces that have already processed and encoded the world according to the logic of the apparatus that produced them — suggests that LegibleArchive is never a neutral achievement: every indexing system, every pathway through an archive, is itself a technical image of sorts, already encoding a particular logic of relevance that shapes what becomes visible and what remains buried regardless of its presence in the archive. Dourish's account of "the stuff of bits" — his insistence that digital information is never purely abstract but is always instantiated in particular material configurations that shape what can be done with it — grounds LegibleArchive in the material reality of infrastructure: an archive's legibility is not just a matter of conceptual organization but of the actual technical systems — search indexes, file formats, server architectures — through which that organization is made operative, systems that themselves age, require maintenance, and can produce their own forms of ArchiveFatigue when neglected. The Socioplastics corpus's gradient system — multiple bibliographies of different sizes, each providing a different entry point into the same underlying field — is an explicit attempt to build LegibleArchive against the ArchiveFatigue that any five-thousand-node corpus inevitably accumulates: not by reducing the corpus, but by multiplying the legitimate pathways into it.