The paper revisits brittle fracture as an energy minimization problem, proposing a variational model for quasistatic crack evolution. Unlike Griffith's theory, which relies on preexisting cracks and well-defined crack paths, this model allows for crack initiation and path quantification. It is numerically implementable in complex settings. The model is based on global energy minimization, where the total energy of the body is the sum of bulk and surface energies. The crack evolves to minimize this total energy. The model is tested on examples, such as a 3-D cylinder under uniaxial tension and a reinforcement tearing in a cylindrical domain. The results show that the model predicts crack initiation and failure under increasing loads, and that the crack path can be determined by minimizing the total energy. The model is compared to Griffith's theory, showing that it can handle both progressive and brutal crack growth, depending on the convexity of the energy function. The paper also discusses the limitations of the model, such as its inability to handle dynamic cases and the need for further research on non-interpenetration conditions. The model is applicable to various settings, including delamination and surface cracks, and is open to non-interpenetration. The paper concludes that the model provides a more comprehensive understanding of brittle fracture than Griffith's theory.The paper revisits brittle fracture as an energy minimization problem, proposing a variational model for quasistatic crack evolution. Unlike Griffith's theory, which relies on preexisting cracks and well-defined crack paths, this model allows for crack initiation and path quantification. It is numerically implementable in complex settings. The model is based on global energy minimization, where the total energy of the body is the sum of bulk and surface energies. The crack evolves to minimize this total energy. The model is tested on examples, such as a 3-D cylinder under uniaxial tension and a reinforcement tearing in a cylindrical domain. The results show that the model predicts crack initiation and failure under increasing loads, and that the crack path can be determined by minimizing the total energy. The model is compared to Griffith's theory, showing that it can handle both progressive and brutal crack growth, depending on the convexity of the energy function. The paper also discusses the limitations of the model, such as its inability to handle dynamic cases and the need for further research on non-interpenetration conditions. The model is applicable to various settings, including delamination and surface cracks, and is open to non-interpenetration. The paper concludes that the model provides a more comprehensive understanding of brittle fracture than Griffith's theory.