This paper proposes an effective method for representing the temporal evolution of legal norms at the component level. While existing frameworks and standards, such as IFLA LRMoo and Akoma Ntoso, provide general tools, a dedicated pattern for fine-grained versioning is needed to reconstruct legal texts at specific points in time, which is essential for reliable AI applications. This paper presents a temporal modeling pattern based on the LRMoo ontology, modeling the evolution of a norm as a diacronic chain of F2 representations. It highlights a key distinction between language-independent temporal versions (TVs)—semantic snapshots of the norm's structure—and linguistic versions (LVs), which are concrete, monolingual implementations. Both versions are modeled as F2 representations linked by the R76 (is derivative of) property. This model is applied recursively to represent the internal structure of a legal text as a parallel hierarchy of abstract component operations (F1 operations) and their versioned component representations (F2 representations). Furthermore, the revision process is formalized using F28 representation creation events, enabling the tracking of changes from specific provisions of an amended law to the precise impact of the amended norm. A case study of the Brazilian Federal Constitution demonstrates that this granular, event-driven architecture can accurately and deterministically retrieve and reconstruct any portion of legal text from a specific date. This model provides a powerful foundation for building verifiable knowledge graphs and advanced AI tools, overcoming the limitations of current generative models.
