It is now over 30 years since the seminal paper of Ivanyi and Lehner (106) suggested a fundamental role for T cells in periodontal disease. It is well established that the development of gingivitis is identical to the development of a delayed-type hypersensitivity reaction and that the so-called stable lesion is essentially a T-cell-mediated response. On the other hand, numerous studies have shown that periodontitis is predominantly a B-cell response with T cells having an immunoregulatory role. Recently however it has been shown that, albeit in mice, the periodontopathic organism Porphyromonas gingivalis down-regulates well over a thousand genes in CD4-positive T cells while up-regulating only about 30 genes. The role of autoimmunity has also been revived with the demonstration of both anti-collagen and anti-heat-shock protein 60-reactive T cells in the gingival tissues. In addition, the so-called T regulatory or T reg cells have been demonstrated in periodontal tissues. These cells are thought to have a central role in the control of autoimmunity. The question therefore arises as to the 'role' of T cells in periodontal disease. This article will explore the concept that because P. gingivalis is turning off T-cell genes it may be the mechanism that enables a balance or stability to be reached between the plaque biofilm and the host, i.e. the 'role' of T cells may be homeostatic rather than defensive or destructive. With respect to autoimmunity in periodontal disease the article will put forward the concept that this is an important and normal component of chronic inflammation. A possible scenario in this context could be that because chronic inflammation is defined as the simultaneous presence of destruction and repair, as the tissue is destroyed, i.e. the collagen is broken down by the various proteases, it must then be removed and the amino acids must be made available for re-use by the fibroblasts. The production of anti-collagen antibodies could then opsonize the broken down collagen fragments and enhance their phagocytosis by fibroblasts, hence facilitating the 're-cycling' process. Such a scenario would suggest that the 'role' of autoimmunity and therefore the controlling mechanisms are an essential part of homeostasis. Similarly, anti-heat-shock protein reactions in chronic inflammation could be the means of removing damaged or destroyed cells locally and again the 'role' of T cells is homeostasis. © 2007 The Authors.