There is convincing evidence that neurological relapses in multiple sclerosis (MS) are the clinical counterpart of acute focal inflammation of the central nervous system (CNS) whereas neurological progression is that of chronic diffuse neurodegeneration. The classical view is to consider that MS is an organ-specific autoimmune disease, i.e. that inflammation is the cause of the neurodegeneration. The succession of relapses eventually leads to accumulation of disability and clinical progression results from subclinical relapses. A series of recent observations tends to challenge this classical concept. Important observations have come from the study of the natural history of MS. In the Lyon MS cohort, accumulation of irreversible disability appeared not to be affected by clinically detectable neurological relapses. This has also been shown to be "amnesic" for the early clinical characteristics of the disease, and essentially age-dependent. Suppressing relapses by disease-modifying agents does not dramatically influence the progression of irreversible disability. Interferons beta reduce the relapse rate by 30% and conventional MRI activity by more than 50%. In spite of this effect on inflammation, the effect on disability is only marginal and possibly relapse-reduction-dependent. Administration of Campath-1H to patients with very active disease in terms of frequency of relapses, accumulation of disability and MRI activity, results in a profound, prolonged lymphopenia and the suppression of clinical and MRI activity, but in spite of this, clinical disability and cerebral atrophy still progress. The same experience has been reported with cladribine and autologous haematopoietic stem cell transplantation. All these observations give support to the fact that relapses do not essentially influence irreversible disability in the long term in MS. They are consistent with what has been shown at the individual level in the 1970s by performing serial quantitative neurological examinations over several years, and with what is currently emerging from early and serial structural brain MRI studies. These breakthroughs have immediate implications for the counselling of patients with MS. They suggest that MS is as much neurodegenerative as inflammatory, and should cause the modification of disease-modifying therapeutic strategies by focussing on the protection and repair of the nervous system and not only on the control of inflammation.