Conceiving of multiple sclerosis directly as a clinical disease entity implies that there are criteria by which anybody having sound experience in neurology would be capable of incontestably diagnosing any instance of multiple sclerosis -- by no more than exploring the individual illnesses' immediate functional repercussions. In other words: Each and every victim of the disease ought to show signs or symptoms which are, as to their nature or course, proper to multiple sclerosis, and which do not occur otherwise.

Clinical criteria have in fact been widely propagated and adopted as the quintessential and ultimate standards for identifying multiple sclerosis, without, however, answering this critical question: What ensures that multiple sclerosis as it is diagnosed from neurological data, and multiple sclerosis as it is identified at post mortem, are truly one and the same? Or, more concretely: How have the clinical and post mortem evidence for the presence of multiple sclerosis been shown to reliably indicate the same kind of damage?

Since multiple sclerosis was originally defined plainly as a macroscopically distinct kind of injury, the neurologist's secondary evidence for the presence of multiple sclerosis should normally have been derived from the condition's genuine macropathological specification. Arguing for the establishment of multiple sclerosis as a clinical disease entity should also have answered the question: Do the unmistakable lesion patterns of multiple sclerosis, or some of the condition's specific traits, in fact necessarily come to be expressed in equally distinctive clinical and, in particular, neurological phenomena? As long as this point has not been clarified, the criteria for the clinical diagnosis of multiple sclerosis can hardly be considered as anything but preliminary and ancillary.

In this connection, it is worth remembering that, if we increase the number of properties required for a particular condition's identification, its specification has not been improved but, quite to the contrary, its definition has been worsened in the form of a superfluous expansion. As to every notion about defining, classifying, or (re-)naming the specific pathology which Carswell first referred to as a "peculiar diseased state" and which then, under Charcot, came to be renamed as "multiple sclerosis", it has thus to be asked whether this particular notion really improves on the genuine lesion's specification -- i.e. whether it makes the condition's characterization more precise, more comprehensive, or more explicit.

It seems accordingly of paramount importance to find out how sound the conception of clinical multiple sclerosis, respectively how reliable the condition's identification by the registration of the patient's history and tracing of his or her neurological dysfunctions ultimately is. To clarify this point, we will now

1. re-evaluate the clinical accounts on the most reliably identified instances of multiple sclerosis to check the diagnostic impact of the particular disease manifestations, and
2. reconsider the reasoning behind the diverse attempts at defining multiple sclerosis on the basis of neurological evidence.